early detection of lung cancer: clinical perspectives of ... · manifest lung cancer toward...

Review

Early Detection of Lung Cancer: Clinical Perspectives of RecentAdvances in Biology and Radiology1

Fred R. Hirsch,2 Wilbur A. Franklin,Adi F. Gazdar, and Paul A. Bunn, Jr.Lung Cancer Program and Departments of Medicine and Pathology.University of Colorado Cancer Center, Denver, Colorado 80262[F. R. H., W. A. F., P. A. B.]; Department of Pathology, University ofTexas, Southwestern Medical Center, Dallas, Texas [A. F. G.]; andDepartment of Oncology, Finsen Center, National UniversityHospital, Copenhagen, Denmark [F. R. H.]

AbstractLung cancer is the most common cause of cancer death

in developed countries. The prognosis is poor, with less than15% of patients surviving 5 years after diagnosis. The poorprognosis is attributable to lack of efficient diagnostic meth-ods for early detection and lack of successful treatment formetastatic disease. Most patients (>75%) present with stageIII or IV disease and are rarely curable with current ther-apies. Within the last decade, rapid advances in molecularbiology, pathology, bronchology, and radiology have pro-vided a rational basis for improving outcome. These ad-vancements have led to a better documentation of morpho-logical changes in the bronchial epithelium before developmentof clinical evident invasive carcinomas. This has changed ourconcept of lung carcinogenesis and emphasized the multistepcarcinogenesis approach on several levels. Combined withthe technical developments in bronchoscopic techniques,e.g.,laser-induced fluorescence endoscope (LIFE) bronchos-copy, we now have improved methods to localize preinvasiveand early-invasive bronchial lesions. With the LIFE bron-choscope, a new morphological entity (angiogenic squamousdysplasia) has been recognized, which might be an impor-tant biomarker and target for antiangiogenic chemopreven-tive agents. To reduce the mortality of lung cancer, thesenew technologies have been taken into the clinic in differentscientific settings. The use of low-dose spiral computed to-mography in the screening of a high-risk population hasdemonstrated the possibility of diagnosing small peripheraltumors that are not seen on conventional X-ray. A shift inthe therapeutic paradigm from targeting advanced clinically

manifest lung cancer toward asymptomatic preinvasive andearly-invasive cancer is occurring. The present article re-views the recent advances in the diagnosis of preinvasive andearly-invasive cancer to identify biomarkers for early detec-tion of lung cancer and for chemoprevention studies.

IntroductionLung cancer is the most common cause of cancer deaths in

the countries of North America and other developed countries,accounting for 29% of all cancer deaths and more deaths thanfrom prostate, breast, and colorectal cancer combined in theUnited States (1). Lung cancer will be diagnosed in;170,000new patients in the United States in the year 2000, and,15% ofthem will survive 5 years after diagnosis (1). The prognosis forthe patients with lung cancer is strongly correlated to the stageof the disease at the time of diagnosis. Whereas patients withclinical stage IA disease have a 5-year survival of about 60%,the clinical stage II-IV disease 5-year survival rate ranges from40% to less than 5% (2). Over two-thirds of the patients haveregional lymph-node involvement or distant disease at the timeof presentation (3). The poor prognosis is largely attributable tothe lack of effective early detection methods and the inability tocure metastatic disease. The unsatisfactory cure rates supportsefforts aimed at early identification and intervention in lungcancer.

Historically, the only diagnostic tests available for thedetection of lung cancer in its early stages were chest radiogra-phy and sputum cytology. The efficacy of these tests as massscreening tools was evaluated in controlled trials sponsored bythe NCI3 and conducted at Johns Hopkins University, MemorialSloan-Kettering Cancer Center, and the Mayo Clinic during the1970s (4–6). The principal goal of these studies was to deter-mine whether a reduction in lung cancer mortality could beachieved by adding sputum cytology testing to annual screeningby chest radiography. Results from these trials showed that bothtests could detect presymptomatic, early-stage carcinoma, par-ticularly of squamous cell type. Resectability and survival rateswere found to be generally higher in the study groups than in thecontrol groups. However, improvements in resectability andsurvival did not lead to a reduction in overall lung cancermortality, the most critical end point. A subsequent study of6346 Czechoslovakian male smokers also found no reduction inlung cancer mortality after dual screening by chest radiographyReceived 6/29/00; revised 10/16/00; accepted 10/30/00.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisementin accordance with 18 U.S.C. Section 1734 solely toindicate this fact.1 Supported by National Cancer Institute Grants CA 58187 from theSpecialized Program of Research Excellence (SPORE)-Lung and CA85070 from the Lung Cancer Biomarkers and Chemoprevention Con-sortium.2 To whom requests for reprints should be addressed, at University ofColorado Cancer Center, Department of Pathology. University of Col-orado Health Sciences Center, 4200 East Ninth Avenue, B 216, Denver,Colorado 80262. E-mail: [email protected].

3 The abbreviations used are: NCI, National Cancer Institute; CIS,carcinomain situ; CT, computed tomography; ASD, angiogenic squa-mous dysplasia; TSG, tumor suppressor gene; LOH, loss of heterozy-gosity; hnRNP, heterogeneous nuclear ribonucleoprotein; SPLC, secondprimary lung cancer; BAL, bronchoalveolar lavage; SCLC, small celllung carcinoma; WLB, white light bronchoscopy; LIFE, laser-inducedfluorescence endoscope; ELCAP, Early Lung Cancer Action Project;PET, positron emission tomography; FDG, [18F]fluoro-2-deoxyglucose.

5Vol. 7 5–22, January 2001 Clinical Cancer Research

Research. on June 12, 2020. © 2001 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

and sputum cytology (7). The negative results from thesescreening studies lead the NCI and other health policy andresearch groups to conclude that mass screening programs in-volving periodic sputum cytological evaluation and chest radio-graphs could not be justified. However, controversies in themethodology and interpretation of the data from these studieshave later been extensively discussed (8, 9). One additionalstudy of annual chest X-ray screening is currently being con-ducted by the NCI; The Prostate-, Lung-, Colorectal-, and Ovar-ian (PLCO) screening trial. This trial includes individuals 55–74years old, but they are not selected for this trial on the basis ofhigh risk for lung cancer (e.g.,smoking history with.20pack-years).

The failure of clinical trials to demonstrate the efficacy ofsputum cytology and chest radiography as mass screening toolshas resulted in a search for better diagnostic approaches forearly lung cancer detection that take advantage of recent devel-opments in molecular biology, gene technology, and radiology(10). Furthermore, as has been the case for mammographyscreening for breast cancer, it has also been important to identifyrisk groups for lung cancer.

Although, much is known about predisposing factors, nat-ural history, and the outcome based on histology and stage, ourunderstanding remains very incomplete in many areas. What arethe early premalignant changes molecularly, biochemically, andmorphologically? Which changes are reversible and which arenot? What research tools are available to provide answers tothese questions? The identification of preinvasive lesions allowsfor developing promising methods for early intervention (11).The therapeutic paradigm and focus are today shifting fromtargeting only clinically verified lung cancer as previously to-ward targeting the premalignant and early- malignant lesions.Furthermore, the prospect of lung cancer screening has todaybecome more meaningful as a consequence of recent develop-ments in biology and radiology and better possibilities to definehigh-risk populations most suitable for lung cancer screening(12).

The present article will focus on the clinical perspectives ofour biological knowledge of premalignant and early-malignantlesions and the potential of the recent technological advance-ment for early diagnosis of lung cancer.

Pathology of Preinvasive and Early InvasiveBronchial Lesions

Most of the efforts to classify lung cancer have beendirected toward invasive carcinoma (13). However, better un-derstanding of the pathogenesis of lung cancer aroused renewedinterest in morphological abnormalities that fall short of inva-sive carcinoma but may indicate initiation of carcinogenesis.These morphological abnormalities are referred to as preinva-sive lesions and are shown in Fig. 1. The last edition of theWHO classification of lung tumors included the classification ofpreinvasive lesions as a separate section. Numerous recent stud-ies have indicated that lung cancer is not the result of a suddentransforming event in the bronchial epithelium but a multistepprocess in which gradually accruing sequential genetic andcellular changes result in the formation of an invasive (i.e.,malignant) tumor. Mucosal changes in the large airways that

may precede or accompany invasive squamous carcinoma in-clude hyperplasia, metaplasia, dysplasia, and CIS (14). Hyper-plasia of the bronchial epithelium and squamous metaplasiahave generally been considered reversible, and not premalignantin the sense of squamous dysplasia and CIS (15).

Squamous metaplasia is a common finding, especially as aresponse to cigarette smoking. Peterset al. (16) studied bron-choscopic biopsies from six sites in 106 heavy cigarette smok-ers; Squamous metaplasia was noted at one or more biopsy sitesin approximately two- thirds of the group, and one-fourthshowed squamous metaplasia in three or more biopsy sites. Theincidence of squamous metaplasia increased with smoking his-tory and was highest in individuals who had smoked more thantwo packs of cigarettes a day. Auerbachet al. (17) noted similarfindings in autopsy tissues: basal cell hyperplasia and squamousmetaplasia are increased in smokers in proportion to smokinghistory. Hyperplasia and metaplasia are believed to be reactivechanges in the bronchial epithelium, as opposed to true preneo-plastic changes (17, 18). The reasons for this include: (a) theyare frequently found in association with chronic inflammation,and may be induced by mechanical trauma; (b) they spontane-ously regress after smoking cessation; (c) in chronic smokers,the molecular changes present in these lesions are similar tothose present in histologically normal epithelium; and (d) thereare no reports linking their presence to increased risk for devel-oping lung cancer. In contrast, moderate-to-severe dysplasia andCIS lesions seldom regress after smoking cessation (19).

Dysplasia and CIS are changes that frequently precedesquamous cell carcinoma of the lung. Saccomannoet al. (20)studied more than 50,000 samples from 6,000 men, many ofwhom had worked in the uranium mining industry. Both smok-ing and uranium mining (radon exposure) were found to beassociated with increased incidence of dysplasia, CIS, and in-vasive cancer. The studies of Saccomannoet al.established thatincreasing degrees of sputum atypia may be recognized anaverage of 4–5 years before the development of frank lungcarcinoma.

Another question is: which grades of sputum atypia pro-gress to cancer? From the Johns Hopkins cohort of the NCIchest X-ray/sputum screening trial, we know that among indi-viduals with moderate atypia on sputum screening,;10% de-veloped known cancer up to 9 years later. Among individualswith severe atypia on the sputum screening,.40% developedknown cancer during the same time period (21). Although thereare data in the literature showing the relationship betweensputum atypia and subsequent invasive cancer, there is still verylittle information about the histological progression in the bron-chial mucosa in the high risk populations. In a recent publica-tion, nine patients with CIS were followed with autofluores-cence bronchoscopy at regular intervals, and 5 (56%) hadprogression to invasive cancer despite endobronchial therapy(22). The number of invasive cancers might even have beenhigher if treatment had not been not given. Ongoing studies ofhigh-risk subjects (e.g.,the Colorado sputum cohort study)including serial follow-up bronchoscopies will provide evidencerelated to the frequency of development of invasive lung canceras it relates to smoking history, airflow obstruction, and sputumatypia.

Since the previous WHO-classification was published in

6 Review: Advances in Early Detection of Lung Cancer



Fig. 1 A, squamous metaplasia. Thecells are widely dispersed, with a reg-ular maturation from the basal regionto the top. There is keratinization, andthe nuclei/cytoplasmic ratio is low.B,moderate dysplasia with ASD. Hyper-cellularity of the epithelium with in-complete maturation and micropapil-lary invasion of capillaries are seen.The nuclei/cytoplasmic ratio is high.C, severe dysplasia. There is markedpleomorphism of the cells with irregu-larity and prominent nucleoli.

7Clinical Cancer Research



1981, two nonsquamous lesions have been added to the WHOclassification of premalignant lesions: atypical alveolar hyper-plasia and diffuse idiopathic neuroendocrine cell hyperplasia(13). Both of these lesions are diagnosed rarely. The formerconsists of lesions,5 mm in diameter and composed of aperipheral epithelial cell proliferation with minimal cytologicalatypia or stromal response and resembles bronchioloalveolarcarcinoma. The lesion has been seen in lung specimens resectedfor lung cancer, but no prospective significance of this lesionhas been reported. However, this morphological lesion may playa role for the pathogenesis of peripheral lung adenocarcinomas(23, 24). The resolution of spiral CT (currently about 3 mm)approaches the diameter of these lesions, and it is anticipatedthat atypical alveolar hyperplasia will be increasingly encoun-tered in subjects undergoing this procedure (25). Diffuse idio-pathic neuroendocrine cell hyperplasia consists of a patchyincrease in the number of well-differentiated neuroendocrinecells in the bronchioles. This process may result in the formationof small carcinoid tumors, and for this reason it is considered“preinvasive.” To date, small cell carcinomas have not beenassociated with this lesion (13).

Recently, the use of fluorescence bronchoscopy (see be-low) has increased the recognition of dysplastic lesions in thelarge airways and a new morphological entity, ASD, was iden-tified (26). Dysplasia of bronchial epithelium in “micropapillo-matosis” and the possible link between angiogenesis and prein-vasive bronchial epithelial dysplasia were recognized as early as1983 by Muller and Muller (27), who also described the ultra-structure of these lesions. It has been suggested that this angio-genesis, which is recognized as capillary loops projecting intothe dysplastic bronchial lining, is responsible for the reducedfluorescence seen in dysplastic lesions by LIFE bronchoscopes(Figs. 1 and 3; Ref. 26). Future prospective studies will showwhether this morphological entity is correlated with a progres-sion to lung cancer so as to be a target for the use of antiangio-genic agents for chemoprevention.

In general, there are several questions/problems relating topremalignant lesions, which will be addressed in future studies:

(a) The morphological criteria for premalignant and early-malignant changes, both on sputum cytology and in bronchialbiopsies, have to be validated for intra- and interobserver repro-ducibility.

(b) Uniform and reproducible morphological/cytologicalcriteria have to be published more extensively, and a training setof slides should be available. By the use of Internet technology,this could be more easily facilitated (28).

(c) The correlation of sputum atypia and histologicalchanges in the bronchi in high-risk population is not welldefined.

(d) The natural course of preinvasive changes in the bron-chi from the high risk subjects needs to be clarified throughlongitudinal, prospective studies with reference to histologicalchanges in the bronchi. Ongoing longitudinal studies with flu-orescence bronchoscopy and multiple biopsies with histologyand other biomarkers will define the ability of these markers toassess for risk.

(e) What is the pathology/biology of the small, often pe-ripherally located, tumors (3 mm in diameter), which are more

often diagnosed with newer radiological techniques (e.g., low-dose spiral CT)?

(f) Optimization of the tissue procurement and processingtechniques are important. Distinction of reactive from neoplasticprocesses is usually straightforward, but diagnostic difficultiesmay arise in the case of (a) inadequate or poorly preparedhistological material to evaluate and (b) the presence of cyto-logical atypia in epithelium stimulated by inflammation, viralinfection, radiation, or chemotherapy.

(g) DNA array analyses of gene expression: will it beuseful? How to collect proper mRNA? Can mRNA extractedfrom microdissected cells obtained at bronchoscopy be globallyamplified and still remain representative of RNA presentin situ?

Biology of Lung Carcinogenesis and Potential EarlyDetection Markers

Lung cancer is the end-stage of multiple-step carcinogen-esis, in most cases driven by genetic and epigenetic damagecaused by chronic exposure to tobacco carcinogens. The geneticinstability in human cancers appears to exist at two levels: at thechromosomal level, including large scale losses and gains; andat the nucleotide level including single or several base changes(29). Lung cancers harbor many numerical chromosomal abnor-malities (aneuploidy) and structural cytogenetic abnormalitiesincluding deletions and nonreciprocal translocations (30). Atleast three classes of cellular genes are involved: proto-onco-genes, TSGs, and DNA repair genes. Oncogenic activation oftenoccurs via point mutations, gene amplification, or chromosomalrearrangement, whereas TSGs are classically inactivated by theloss of one parental allele combined with a point or smallmutation or aberrant methylation of a target TSG in the remain-ing allele. Additionally, dysregulated gene expression (eitherincreased or decreased expression) can occur by other, as yetunknown, mechanisms (30). Present studies have not yet con-firmed a prominent role for abnormalities of DNA repair genesin lung cancer.

Preneoplastic cells contain several molecular genetic ab-normalities identical to some of the abnormalities found in overtlung cancer cells (Fig. 2). These include allele loss at severalloci (3p,9p,8p, and17p),mycandras up-regulation, cyclin D1overexpression, p53 mutations, and increased immunoreactiv-ity, bcl-2 overexpression and DNA aneuploidy (31–35). Allelo-typing of precisely microdissected, preneoplastic foci of cellssuggests that the earliest changes in the bronchial epithelium isallele loss at chromosome regions 3p, then 9p, 8p, 17p, 5q, andthen ras mutations (36–39).The biological meaning of LOH isonly vaguely understood. Recent evidence suggests that LOHmay be a consequence of mitotic recombination, that there isonly infrequent physical loss of genetic loci, and that LOHprobably precedes chromosomal duplication (40). Allelic losswould thus be significant primarily in the presence of mutationin the retained allele, and gene dosage would not be expected toexert a phenotypic effect in LOH. Some reports have indicatedthat ras activation occurs at early carcinoma stages (34). His-tologically normal bronchial epithelium adjacent to cancers hasalso been shown to have certain genetic losses. Atypical ade-nomatous hyperplasia, the potential precursor lesion of adeno-carcinomas, often have Ki-rasmutations (41).




Molecular changes have been found not only in the lungsof patients with lung cancer, but also in the lungs of current andformer smokers without lung cancer (18, 42, 43). These obser-vations are consistent with the multistep model of carcinogen-esis and “field cancerization” process, whereby the whole regionis repeatedly exposed to carcinogenic damage (tobacco smoke)and is at risk for developing multiple, separate, clonally unre-lated foci of neoplasia. The widespread aneuploidy that occursthroughout the respiratory tree of smokers supports this theory(44). However, the presence of the same somatic p53 pointmutation at widely dispersed preneoplastic lesions in a smokerwithout invasive lung cancer indicates that expansion of a singleprogenitor clone may spread throughout the respiratory tree(45). These molecular alterations might thus be importanttargets for use in the early detection of 1ung cancer and for useas surrogate biomarkers in the follow-up of chemoprevention

studies. Detection of these mutant cells should be possible withthe different molecular techniques in accessible specimens. Theprospects of diagnosing these biological abnormalities in mul-tiple types of clinical specimens are discussed below.

Specimens for Clinical Testing: SputumSince the 1930s, cytological examination of sputum has

been used for the diagnosis of lung cancer (46). Cytologicalexamination of sputa, especially multiple samples, is helpful forthe detection of central tumors arising from the larger bronchi(e.g., squamous cell- and small cell carcinomas). Exfoliatedcells from peripheral tumors, such as adenocarcinomas, arisingfrom the smaller airways (small bronchi, bronchioles, and alve-oli), especially those less than 2 cm in diameter, can be detectedonly occasionally in sputum samples. This has become ofgreater importance because the changes in cigarette exposition

Fig. 2 Sequential changes during lung cancer pathogenesis. Although multiple genetic markers are abnormal in lung cancers, their appearance duringthe lengthy preneoplastic process varies. The timing of the appearance of these changes has been investigated in bronchial preneoplasia, becausesequential sampling of the peripheral lung is technically difficult. Several alterations have been described in histologically normal bronchial epitheliumof smokers. Other changes have been detected in slightly abnormal epithelium (hyperplasia, metaplasia) which we do not consider to be truepremalignant lesions. These changes are regarded as early changes. Molecular changes detected frequently in dysplasia are regarded as intermediatein timing, whereas those usually detected at the CIS or invasive stages are regarded as late changes. It should be stressed that although there is a usualorder, exceptions regarding the timing of onset may occur. Some changes are progressive, such as chromosome 3p deletions. Thus small discretechanges are present early, progressively become more extensive during pathogenesis, and frequently involve all or almost all of the arm in CISsamples. Although allelic loss at the TP53 locus may precede the onset of mutations, data on this sequence are scant. Dysregulation of the RNAcomponent of telomerase (with its appearance in nonbasal cells) is an early event, whereas up-regulation of the gene is a relatively late event.




(filters and decreased nicotine content) have created an increasein adenocarcinomas and a decrease in squamous carcinomas(47–49). The sensitivity of sputum cytology for early lungcancer is only in the 20%–30% range from screening studies,but by adhering to proper specimen collection, and processingand interpreting criteria, the yield can be substantially improved(50, 51). The data on the reliability of the sputum are conflicting(52–54). Browmanet al. (52) reported interobserver agreementof 68% for exact and 82% for within - 1–category. Hollidayetal. (54) reported low agreement within observers (27–60%) andacross observers (13–50%). Within - 1 - category intraobserveragreement underwent a two- or 3-fold increase in agreement,which was also the case for interobserver agreement. The var-iation in intra- and interobserver agreement seems to depend onexperience among the cytotechnicians/cytopathologists and thecomposition of categories studied. A higher degree of agreementis obtained for higher grades of dysplasia (54). Risseet al. (55)showed that the ability to detect premalignant conditions isdependent on the number and type of cells present in the deeperairways, suggesting a mode of improvement that is unrelated toobserver reliability. MacDougallet al. (56) concluded that spu-tum cytology was too insensitive and insufficiently accurate tobe included in the routine work-up of any patient suspected ofhaving lung cancer. To improve the reliability of sputum cytol-ogy examinations a simplification of the diagnostic categoriesfrom 6 (normal; squamous metaplasia; mild, moderate, andsevere atypia; and carcinoma) to 2–3 categories have beenproposed (54). Future clinicopathological studies will be re-quired to validate this concept.

To improve the sensitivity of sputum examination as apopulation-screening tool for the detection of early lung cancer,several approaches are currently under development.

Immunostaining. Annual sputum specimens obtainedfrom individuals screened at Johns Hopkins were obtained, andthe patients were monitored for 8 years (57). Because theclinical outcome of these patients was known, archival sputumspecimens were screened for the presence of biomarkers thatcould indicate the presence of lung tumors in an early, preinva-sive stage. In an attempt to distinguish the pattern of markerexpression Tockmanet al. (58) studied two monoclonal anti-bodies. Positive staining predicted subsequent lung cancer ap-proximately 2 years before clinical recognition of the disease,with a sensitivity of 91% and a specificity of 88% (58). One ofthese antibodies (703 D4) had a higher sensitivity and was lateridentified as recognizing hnRNP A2/B1 (59). The role ofhnRNP A2/B1 overexpression for detecting preclinical lungcancer has been studied in a large high-risk population including6000 Chinese tin miners who were heavy smokers and who hadan extraordinary rate of lung cancer (60). The results from thisstudy indicated that detection of hnRNP A2/B1 overexpressionin sputum epithelium cells was 2- to 3-fold more sensitive fordetection of lung cancer than standard chest X-ray and sputumcytology methods. The method was particularly effective inidentifying early disease (60). The sensitivity was 74%versus21% for cytology and 42% for chest X-ray. However, thebiomarker had a lower specificity (70%) compared with cytol-ogy (100%) and chest radiograph (90%). An ongoing clinicaltrial is evaluating the performance of the A2/B1 protein as abiomarker for the early detection of SPLC. The patients at risk

for SPLC have the highest incidence of lung cancer (2–5%)among asymptomatic populations (61). In this trial, 13 SPLCswere identified by A2/B1, and the sensitivity and specificitywere 77–82% and 65–81%, respectively. Among the casesidentified as positive by immunocytochemistry and image cy-tometry, 67% developed SPLC within 1 year (62). Whereas theprevious immunocytochemistry studies on material from theolder screening material from the NCI-supported screeningstudies were made on sputum cells cytologically classified withmoderately or gravely atypical metaplastic appearance, the latterstudies have been done on cytologically “normal appearing”cells. More recently Sueokaet al.(63) reported the confirmationof the value of overexpression of hnRNP A2/B1 to detectpreclinical lung cancer in Japan. Efforts to improve the sensi-tivity of hnRNP markers are ongoing (64).

PCR Techniques. PCR techniques have been used forthe evaluation of molecular biomarkers for early lung cancerdetection. In a pilot study with selected patients from the JohnsHopkins Lung Project (JHLP), 8 (53%) of 15 patients withadenocarcinoma or large cell carcinoma were detected by mu-tations in sputum cells from 1 to 13 months before clinicaldiagnosis (65). However, the method seemed to be less sensitivethan the protein marker described above, and the identificationof specific gene abnormalities is further limited by the need toknow the specific mutation sequence with which to probe thesputum specimens. Currently, this approach is not practical forscreening undiagnosed individuals. Future advances in genechip technology may permit testing for all possible mutations ofcommon oncogenes and TSGs in clinical specimens of asymp-tomatic individuals (62).

Microsatellite markers are small repeating DNA sequencesfound in the noncoding regions of a gene. PCR amplification ofthese repeat sequences provides a rapid method for assessmentof LOH and facilitates the mapping of suppressor genes (66,67). Microsatellite alterations are extension or deletions of theserepeated elements. Detection of microsatellite alterations inhistological or cytological specimens may facilitate the detec-tion of clonal preneoplastic or neoplastic cell populations. Al-though the detection of microsatellite alterations does not indi-cate the specific genetic change in the tumor, detection of clonalcell populations might serve as a cancer screening marker (65).Identical alterations have been found in lung cancers and cor-responding sputum samples demonstrating minimal atypia (68).The p16 gene is located on the short arm of chromosome9(9p21) and is frequently mutated or inactivated in tumors andcell lines derived from lung cancer (69, 70). Belinskyet al. (71)measured hypermethylation of the CpG islands in the sputum oflung cancer patients and demonstrated a high correlation withearly stages of non-small cell lung cancer, which indicated thatp16 CpG hypermethylation could be useful in the prediction offuture lung cancer. However, prospective studies are needed toevaluate the role of p16 hypermethylation as a marker for earlylung cancer detection. Multiple other genes are inactivated byhypermethylation in lung cancer (72), and the detection ofhypermethylation may be useful for risk assessment and earlydiagnosis.

Computer-assisted Image Analysis. Computer-assistedimage analysis was initially used to detect malignancy-associ-ated changes (e.g.,subvisual or nonobvious changes in the




distribution of DNA in the nuclei of histologically normal cellsin the vicinity of preinvasive or invasive cancer; 73). In aretrospective analysis of sputum cytology slides, malignancy-associated changes alone correctly identified 74% of the sub-jects who later developed squamous cell carcinoma (74). Thetechnique has been improved, and recent data showed sensitiv-ities of 75% for stage 0/I lung cancer and 85% for adenocarci-nomas with a specificity of 90% (75). This quantitative micros-copy technique allows the examining of thousands of cells perslide within a relative short time. Similar techniques have beenapproved in the United States for cervical cancer screening, andmight, in the future, play a role for lung cancer screening.However, no prospective clinical studies has evaluated thistechnology in a larger lung cancer screening setting.

High Throughput Technology. With future advances ingene chip technology, it might become feasible to probe forexpression of multiple genes in sputum specimens of asymp-tomatic individuals. However, this requires a large amount ofundegraded RNA from respiratory tract cells. With the highthroughput technology, a higher sensitivity might be achievedby using multiple markers at the cost of achieving a lowerspecificity, which would be undesirable for a screening study.

In conclusion, we need to reevaluate the role of sputumcytology for screening and early detection of lung cancer be-cause of advances in biomarkers and technology. Ongoing stud-ies with standard and biomarker analysis in high-risk groupsmight change the previous negative attitude and provide a newperspective on sputum cytology as a mass screening tool whenapplied in a high-risk population. Adding different moleculardiagnostic tests gives the possibility for early diagnosis far inadvance of clinical presentation. However, validation of thetests in larger prospective studies is necessary, and the individ-ual tests have to be compared with each other to define the roleof early diagnosis in the overall management of high-risk sub-jects. Furthermore, health economic issues have to be consid-ered.

Specimens for Clinical Testing: BALBAL involves the infusion and reaspiration of a sterile

saline solution in distal segments of the lung via a fiberopticbronchoscope. Ahrendtet al., (76) examined a series of 50resected non-SCLC tumor patients and compared the tumor andBAL with regard to molecular markers including p53 mutations,K-ras mutation, the methylation status of the CpG island of thep16 gene, and microsatellite alteration (Tables 1 and 2). Withthe possible exception of the test for microsatellite alteration, allof the tests had relatively high sensitivity and could detectmutant cells in the presence of a large excess of normal cells.The frequencies of these changes in the tumors ranged from27% (for K-ras mutations) to 56% (for p53 mutations). Asexpected, p53 mutations were more frequent in central (predom-inantly squamous cell) tumors, and K-ras mutations were morefrequent in peripheral (predominantly adenocarcinoma) tumors.The specificity was high (nearly 100%) because, with the ex-ception of microsatellite alterations, the same genetic change inBAL sample as in tumors was always found, but the sensitivitywas low, and in only 53% of tumors that contained molecularlesions were the same abnormalities detected in correspondingBAL fluids. Specifically, the tests were least helpful in the

group of patients in whom improved diagnostic abilities aremost needed, those with small, peripherally located tumors (77).Unfortunately, the investigators were not able to compare themolecular tests with routine cytopathological analysis of theBAL specimens. The sensitivity of the molecular tests in BALspecimens has to be improved, and we need to know the resultsfrom subjects at increased risk (current and former smokerswithout lung cancer and survivors of previous cancer of theupper respiratory tract) and subjects with chronic lung diseasesas well as results from healthy never smokers.

A European group has previously shown that genetic al-terations detected in DNA from bronchial lavage of individualswith lung cancer were also found in individuals with no evi-dence of malignant disease (78), which raises the question aboutthe specificity of such molecular damage in neoplastic condi-tions. To improve the sensitivity and specificity of detectingallelic imbalance in lung tumors, high throughput PCR-basedmicrosatellite assays have been established (79). In a recentstudy by Fieldinget al. (80), the up-regulation of hnRNP A2/B1was found to be a promising marker in BAL for the detection ofpremalignant and malignant bronchial lesions with a diagnosticsensitivity of 96% and a specificity of 82%.

It is too early yet to make conclusions as to whether BALexaminations will add to other pathological/molecular biologi-cal clinical studies. To obtain diagnostic material for BALbronchoscopy is required, and we do not have any data thatcompare BAL examinations with biopsies. Thus, we do notknow whether BAL is a valuable adjunct to the biopsies takenunder the same bronchoscopy procedure.

Specimens for Clinical Testing: Peripheral BloodFor many years scientists have searched for a lung cancer-

specific tumor marker that could be detected in peripheral blood.Optimism was raised in the “early” immunocytochemistry eraby the use of monoclonal antibodies raised against more-or-lessspecific epithelial epitopes. In the search for epithelial cells inperipheral blood and bone marrow, monoclonal antibodiesagainst cytokeratin have been used. However, these reactionsare clearly not cancer-specific, and some antibodies have beenshown to cross-react with normal blood or bone marrow ele-ments (81, 82). Another explanation could be that cells from themacrophage/monocyte system may contain proteins derivedfrom the primary tumor that have undergone necrosis andapoptosis and that these processed proteins are recognized bythe antibodies (82). On the basis of “traditional” immunocyto-chemistry, no markers have been able to detect premalignant orearly-malignant disorders based on a peripheral blood sample.However, with the development of DNA technologies, newpossibilities have been raised, and, with the use of PCR tech-niques, some promising reports have been published.

Nanogram quantities of DNA circulating in blood are pres-ent in healthy individuals (83, 84). Tumor DNA is also releasedinto the plasma component in increased quantities (85, 86).Thus, the plasma and serum of cancer patients is enriched inDNA, an average four times the amount of free DNA as com-pared with normal controls (87). In a study by Chenet al. (88),a comparison of microsatellite alterations in tumor and plasmaDNA was done in SCLC patients, and 93% of the patients with




microsatellite alterations in tumor DNA also had modificationsin the plasma DNA. However, some patients had LOH only inthe tumor DNA. Because most of the microsatellite alterationswere similar in tumor DNA and plasma DNA, they concludedthat some of the DNA circulating in the blood comes from thetumor. Thus, modifications of circulating DNA can be used asan early detection marker. Detection of aberrant DNA methyl-ation in serum DNA in patients with non-SCLC has beenreported (72). Although the number of patients was small andthe hypermethylated DNA was found in all stages, it opens upfor the possibility to be used as an early lung cancer detectionmarker. Furthermore,p53 and ras gene mutations have been

detected in the plasma and serum of patients with colorectalcancers (89–91), pancreatic carcinomas (92, 93), and hemato-logical malignancies (94).

In conclusion, the limited direct accessibility of lung car-cinomas has led to efforts to identify tumor-associated solublemarkers in serum or plasma. Many of the currently recognizedsoluble markers were first identified as “tumor” markers but,when evaluated in nonneoplastic tissue, have often been foundin normal cells as well as in tumors. For early detection of lungcancer, we need more clinical data evaluating these new molec-ular biological markers from multiple sites, especially in high-risk groups.

Table 1 Tissues and other resources for the study of molecular markers

Specimen Ref. Comments

Tumor tissue Numerous Mixture of cell types, may require microdissection (139). Extensively usedfor most studies. Alcohol-fixed fine-needle aspirates may be used formutational and other studies.

Sputum 65, 68, 71, 74 Respiratory cells usually in small minority. Most samples fixed inSaccomanno’s fixative. Several studies have been performed on thesespecimens many years later.

Surrogate organ 140 Predominantly squamous epithelial cells. Buccal smears, brushings oftongue or tonsil may be explored as potential surrogate organs resultingfrom the field effect of tobacco damage of the entire upperaerodigestive tract. This concept needs to be confirmed.

Bronchial brush/wash 141, 142, 143 Predominantly respiratory cells. Fresh, frozen, or alcohol-fixed samples aresuitable for multiple studies including FISH.a

Bronchial tissues 42, 43, 45, 144, 145 Usually from bronchial biopsies, but may be obtained from surgicalresection specimens. Formalin fixation and paraffin embedding requiredfor histological diagnosis, although EASI preps may permitidentification and isolation of subpopulations. Paraffin sections may beused for genotyping polymorphisms, for allelotyping, and forin situhybridization.

BAL fluids 76, 78, 146, 147, 148 BAL fluids are useful for examining the peripheral airway cells, which arethe precursor cells of most adenocarcinomas. Numerous mononuclearcells present. Enrichment of epithelial cells desirable.

Blood components 72, 92, 149 Analysis of circulating tumor cells and genetic material shed by dyingtumor cells into the plasma component may yield useful biological anddiagnostic information. Gene mutations and presence of epithelial cellmarkers have been used to detect circulating tumor cells. Genemutations, allelic loss, microsatellite alterations, and aberrantmethylation have been used to identify tumor cell DNA released intothe fluid compartment.

Tissue for molecular staging 150, 151 Although little data exists for lung cancers, regional lymph nodes, sentinellymph nodes, and surgical resection margins have been used in othertumor types for molecular staging.

Tumor cell lines 152, 153 Provide an unlimited self-replicating source of high-quality molecularreagents and have been used for numerous studies. Cell lines may ormay not reflect the properties of the tumors from which they werederived (26), although they probably represent cellular subpopulations(27). Aggressive metastatic tumors are more likely to be successfullycultured (28) resulting in skewed data.

Cultures of nonmalignant tissues 154, 155 Epithelial cultures may be useful for studying molecular changes duringmultistage pathogenesis. Temporary as well as a few immortalizedcultures from nonmalignant epithelial cells have been established.B-lymphoblastoid cultures are useful for linkage analysis, for geneticsuspectibility studies, and for allelotyping corresponding tumors.

Nonmalignant tissue from patientsand from cancer-free relatives

156, 157, 158 Tissues such as buccal smears, tumor-free lymph nodes, and peripheralblood mononuclear cells are useful as controls for linkage analysis, forgenetic susceptibility studies, and for allelotyping corresponding tumors.

a FISH, fluorescencein situ hybridization; EASI, epithelial aggregate separation and isolation.




Specimens for Clinical Testing: BronchoscopyWLB is the most commonly used diagnostic tool for ob-

taining a definite histological diagnosis of lung cancer. Bron-choscopy has major diagnostic limitations for premalignant le-sions. Because these lesions are only a few cells thick (0.2–1mm) and have a surface diameter of only a few millimeters, theyrarely are observed as visual abnormalities. Woolner (95) re-ported that squamous cell CIS was visible to experienced bron-choscopists in only 29% of cases. To address this limitation,fluorescence bronchoscopy was developed. Early studies offluorescence bronchoscopy entailed the use of fluorescent drugs(hematoporphyrin dyes) that were preferentially retained in ma-lignant tissue (96). Although, studies evaluating this approachdid, in fact, show that early invasive andin situcancers could belocalized, the detection of dysplasia remained problematic (97–100). Furthermore, the development of photodynamic diagnos-tic systems was hampered by problems including skin photo-sensitizing and interference with tissue autofluorescence. Toovercome these problems, a new laser photodynamic diagnosticsystem was developed (101). This system detected tumor-specific drug fluorescence at 630 nm wavelength, which is farfrom normal tissue autofluorescence (500–580 nm), and inter-ference by autofluorescence from normal tissue should thenhave been eliminated, but it remained a significant problem(102).

Another approach was developed by Palcicet al. (103),who noticed the lack of autofluorescence in the tumor lesions byusing blue light (442 nm) rather than white light to illuminatethe bronchial surface. They amplified the difference in autofluo-rescence between normal, premalignant, and tumor tissue forclinical use (103, 104). Using a high-quality-charge coupleddevice and special algorithm, the LIFE was developed, takingadvantage of the principle that dysplastic and malignant tissuesreduce autofluorescent signals compared with normal tissue(Fig. 3).

Several studies have been performed comparing the diag-nostic specificity and sensitivity of LIFE bronchoscopyversusWLB in diagnosing preinvasive and early-invasive lesions(105–109; Table 3). Most of the studies reported a higherdiagnostic sensitivity of LIFE bronchoscopy in the detectionpremalignant and early-malignant lesions at the cost of lowerspecificity (i.e.,more false-positive results). In most of thesestudies, lesions with moderate dysplasia or worse were the targetof the study and rated as “positive.” The prevalence of prein-vasive and early lung cancer varies widely from one study toanother, from 20.2% (105) to 65.8% (102). The explanationmight be beyond the risk profile of genetic variations or differ-ent levels of experience among the endoscopists as well as thepathologists involved. Furthermore, there seems to be a trainingeffect in using the LIFE bronchoscope, which has been demon-

Table 2 Molecular approaches for lung cancer investigation

Specimen Ref. Comment

Gene mutations 159, 160, 161 Widely used technique, especially forp53 and ras genes. Often usedto determine the role of a newly discovered gene in thepathogenesis of lung cancer. May be of diagnostic and prognosticsignificance. Multiple methodologies available.

Allelotyping 18, 158 Useful as a partial substitute for mutational analysis and fordetermining the chromosomal locations of putative tumorsuppressor genes. Widely used to study multistage pathogenesis.Readily performed on formalin-fixed and microdissected tissues.Increasing use of genotyping using automatic sequencers.

Gene expression at RNAand protein level

145, 162, 163, 164, 165, 166 Northern blotting and reverse transcription-PCR are widely used toinvestigate gene expression. Western blotting often used fordetection of protein expression.In situ hybridization for messageexpression can be performed on paraffin-embedded tissues and,thus, can be used to investigate multistage pathogenesis.Microarray techniques offer promise of examining all or most ofthe genome but currently require relatively large amount of high-quality RNA from purified cell populations. Sage technique usefulfor investigation and identification of expressed genes. Similarly,advances in proteomics will permit simultaneous detection ofmultiple proteins. Numerous immunohistochemical studies ofoncogene expression have been used to study multistagepathogenesis. Of particular interest, hnRNP expression onexfoliated epithelial cells in sputum samples may predict fordevelopment of cancer.

Molecular cytogenetics 40, 167, 168, 169, 170 In situ hybridization studies of fixed materials or using smears hasprovided considerable information about numerical and structuralchanges.

Comparative genomichybridization

171, 172 Useful for detection of gene amplifications. Less sensitive for thedetection of regions of allelic loss.

Morphometric studies 74, 173, 174 May be applied to paraffin-embedded tissues. Useful for determininganeuploidy and for measuring a number of nuclear andcytoplasmic parameters.




strated by Venmanset al. (107). In their study, the diagnosticsensitivity increased from 67 to 80% when comparing the firstand the second half of the study. The use of the LIFE device inconjunction with WLB improved the detection rate of preneo-plastic lesions and CIS significantly (Table 3). Kurieet al.(106)looked for more subtle tissue transformation, but their studyincluded few patients with moderate dysplasia or worse. Noimprovement in the evaluation of metaplasia index was ob-served by the use of LIFE bronchoscopy. Thus, differences inthe study population might explain the different conclusion.There are still no clinical studies with sufficient long-term datashowing that moderate dysplasia is the most relevant clinicalpredictor of eventual malignancy. Limitations in making con-clusions from the existing studies are also the potential meth-odological bias related to the order in which the different bron-choscopy procedures are done and whether the same examinerhas performed both procedures. To address these issues, a

prospective randomized study between LIFE bronchoscopy andWLB was done at the University of Colorado Cancer Center.The study design included a randomization with regard to theorder of procedure as well as the order of the individual bron-choscopist (109). The order of the procedure and of the indi-vidual bronchoscopist did not affect the results. The study alsodemonstrated a significantly higher sensitivity in detecting pre-malignant lesions visualized by the LIFE, but at the cost of alower specificity (109). The reason for the low diagnostic spec-ificity found with the LIFE bronchoscopy in the different studiesmight be attributable to the visualization of more abnormal fociwith the LIFE bronchoscope, with the consequence that a largernumber of biopsies were taken and, thus, there was a higher riskof more false-positive results. The use of LIFE bronchoscopyhas led to the identification of a new morphological entity, theASD, which is described above. In a recent morphological studyangiodysplastic changes were frequently found in preneoplastic

Fig. 3 A, normal WLB and normal LIFE bronchoscopy.B, WLB shows inflammatory changes in the bronchial mucosa but no suspicion ofmalignancy (left). LIFE bronchoscopy shows diffuse reduced autofluorescence (visualized bydiffuse red-brownish colorization;arrows). Biopsydemonstrated diffuse severe dysplasia.




and early-malignant lesions in the bronchi (26). The morpho-logical entity has been confirmed in preneoplasias among smok-ers, and the perspectives of this finding have been extensivelydiscussed (110). The prognostic significance of this morpholog-ical entity is currently studied in ongoing long-term follow-upstudies. Future studies have to evaluate the role of ASD as abiomarker for early lesions and whether it can be used as amarker for treatment effect or therapeutic target for chemopre-vention.

The LIFE bronchoscope may play an important role in thescreening and follow-up of subjects at high risk of developinglung cancer. At this stage, however, it is unknown whether theLIFE bronchoscope will lead to a reduction in lung cancermortality. There are also no data on cost-effectiveness andcost-benefit analyses available for this new diagnostic proce-dure. The use of the LIFE bronchoscope may also in the futurebe extended to other indications,e g.,patients staged as havingresectable lung cancer on one side. Whether LIFE bronchoscopyof the contralateral lung will disclose abnormalities, whichwould change the therapeutic decision, is not yet reported.

Recent Advances in RadiologyThe previous NCI-sponsored screening trials failed to dem-

onstrate any reduction in the lung cancer mortality by sputumcytology and yearly chest radiography as mass screening toolsfor lung cancer screening. Limitations of design and executionof the studies, however, have been discussed extensively (8,111, 112). An extended follow-up (median, 20.5 years) of theMayo Lung Project was recently published (113). There wasstill no difference in lung cancer mortality between the inter-vention arm and the control arm (4.4versus3.9 deaths per 1000person-years). However, the median survival for patients withresected early-stage disease was 16.0 years in the interventionarmversus5.0 years in the usual-care arm (P, 0.05). The latterfindings have raised the question as to whether some smalllesions with limited clinical relevance may have been identifiedin the intervention arm, and the question of “overdiagnosis” wasdiscussed in accompanying editorials (114).

Mass screening for lung cancer has been performed inJapan for many years and has been performed in over 500,000people in about 80% of the local communities (115). Sobueetal. (116) observed that annual clinic-based chest X-ray screen-ing for lung cancer in Japan showed reduced lung cancer mor-tality by about one-fourth among individuals who underwentscreening once a year. In this screening program, the relativeodds ratio of dying from lung cancer within 12 months was0.535 and in the 12–24-month period was 0.638 (117). How-ever, many studies have focused on the pitfalls in the detectionof abnormalities by radiography (118–122). The limit of chestradiographic sensitivity for nodule detection is roughly 1 cm indiameter, by which time the tumor has over 109 cells and mayalready have violated bronchial epithelium and vascular epithe-lium. CT has been shown to be more effective in the detectionof peripheral lung lesions compared with plain radiography orconventional tomography of the whole lung (123, 124).

Spiral CT scan is a relatively new technology with theability to continuously acquire data resulting in a shorter scan-ning time, a lower radiation exposure, and improved diagnosticaccuracy compared with those of plain radiography (125–127).Spiral CT allows the whole chest to be imaged in one or twobreath-holds, reducing motion artifacts and eliminating respira-tory misregistration or missing nodules. Although there isgreater radiation exposure with CT than with chest radiography,low-dose techniques (lower mA of 30–50 compared with 200for conventional CT) have achieved calculated exposure dosesthat are 17% that of conventional CT and 10 times that of chestradiographs. Further reduction in radiation dose while maintain-ing diagnostic accuracy is a topic of current research. Further-more, for the baseline screening, low-dose spiral-CT-scan i.v.contrast is not administered. Nodules as small as 1–5 mm can beshown with modern spiral CT technology (25, 128). The obvi-ous advantages with this new technology led some groups inJapan and in the United States to look to low-dose spiral CT asa tool for screening (Refs. 129–131; Tables 4 and 5).

In a Japanese report, spiral CT scans and chest radiographswere done twice a year in 1369 individuals (129). Peripherallung cancer was detected in 15 (0.3%) of 3457 examinations,and, among the 15 lung cancer cases detected, the results ofchest X-ray were negative in 11(73%), and the tumors weredetected only by low-dose spiral CT. The detection rates oflow-dose spiral CT and chest X-ray were 0.43% (15 of 3457examinations) and 0.12% (4 of 3457 examinations), respec-tively. Furthermore, 14 (93%) of the 15 lung cancers were stageI disease. The histology showed that 11 of the 15 lung cancercases were adenocarcinoma, and 4 had squamous cell carci-noma. The effective exposure dose with spiral CT scan in thatstudy was calculated to about one-sixth that of conventional CT.

The ELCAP in New York was designed to determine: (a)the frequency with which nodules were detected; (b) the fre-quency with which detected nodules represent malignant dis-ease; and (c) the frequency with which malignant nodules arecurable (131). In the ELCAP study, 27 lung cancers were foundamong 1000 subjects screened. Among the 27 patients withcancer, 85% had stage I disease (Table 5).

Another population-based study on low-dose CT screeninghas been published by Soneet al. (130), using a mobile low-dose spiral CT scanner. The detection rate was 0.48% (i.e.,4–5

Fig. 4 Seventy-one-year-old man with a spicular nodule in upper leftlobe demonstrated on low-dose helical CT (picture), but not visible onchest X-radiography. CT- guided biopsy showed adenocarcinoma.




cases per 1000 examinations). Surprisingly, there was no dif-ference in the detection rate among smokers (0.52%)versusnonsmokers (0.46%). The results from the three population-based studies are summarized in Tables 4 and 5. The conclusionfrom these studies is that 85% of the lung cancers detected bylow-dose CT were in stage I, offering improved possibility forcurative treatment and better prognosis in general. However, theissue of “false-positive” scans has to be taken into consideration.Thus far, up to 20% of the participants with nodules on the scanhad no malignancy during the follow-up period. The possibilitythat the cancers found represent incidental cancers as in theMayo Lung Project must also be considered (114). The resultsfrom these studies confirm the expectation that low-dose CTincreases the detection of small noncalcified nodules and, thatlung cancer at an earlier and more curable stage are detected.The mobile CT screening study by Soneet al.(130) showed thatlow-dose CT increased the likelihood of detection of malignantdisease 10 times as compared with radiography. The overall rateof malignant disease was lower in the Japanese studies (129,130) compared with the ELCAP study (Ref. 131; detection rates0.43–0.48%versus2.7%). This could be because the Japanesestudies screened individuals from the general population ages

40–74, whereas ELCAP screened people at high risk, ages$60,with a tobacco history of at least 10 pack-years. Thus, asexpected, the risk of the population to be screened affects therate of cancer detection.

Questions remaining to be answered include: (a) what arethe diagnostic sensitivity and specificity of this procedure; and(b) does screening reduce lung cancer mortality? The spiral CThas not been as sensitive for small central cancers as it is forsmall peripheral cancers (129, 131). Minute nodules of lungcancer that are near the threshold of detectability may be over-looked at spiral CT screening (132). A prospective study of thediagnostic sensitivity of spiral CT has recently shown that thediagnostic sensitivity exceeded the sensitivity of conventionalCT in previous reports (25). However, there were limitations inthe detection of intrapulmonary nodules smaller than 6 mm andof pleural lesions. Compared with surgery (thoracotomy withpalpation of deflated lung, resection, and histology), the sensi-tivity of spiral CT was 60% for intrapulmonary nodules of,6mm and 95% for nodules of$6 mm and was 100% for neo-plastic lesions$6 mm. Furthermore, a marked difference in thesensitivities of two independent observers was found for nod-ules smaller than 6 mm, whereas agreement was much better for

Table 3 BronchoscopyversusWLB in diagnosing premalignant and early-malignant lesions

AuthorNo. of

biopsies

Sensitivity

Relativesensitivity

LIFE1WLB

Specificity

Relativespecificity

LIFE1WLB

Predictive values

LIFE1WLB LIFE WLB

LIFE1WLB LIFE WLB

PPVa

LIFE1WLB

NPVLIFE1WLB

PPVLIFE

NPVLIFE

PPVWLB

NPVWLB

Lam et al. (105) 700 0.67 NR 0.25 6.3 (2.7)c 0.66 NR 0.90 NR 0.33 0.89 NR NR 0.39 0.83Kurie et al.b (106) 234 NR 0.38 NR NR NR 0.56 NR NR NR NR 0.16 0.81 NR NRVenmanset al. (107) 139 NR 0.89 0.78 1.43 NR 0.61 0.88 NR 0.20 NR 0.14 0.99 0.32 0.98Vermulenet al. (108) 172 0.93 NR 0.25 3.75 0.21 NR 0.87 NR 0.13 0.96 NR NR 0.19 0.90Kennedyet al. (109) 394 0.79 0.72 0.18 4.4 0.3 0.43 0.78 0.38 0.21 0.85 0.25 0.87 0.17 0.80

a PPV, positive predictive value; NPV, negative predictive value; NR, not reported.b Based on reference pathologist.c If invasive carcinoma is included.

Table 4 Results from three population-based screening studies with low-dose spiral CT (LDCT)

AuthorsNo. of individuals

studiedTrue

positivenFalse

positivea %Predictivevalue %

Detection rate %

Pack-yrAge incl.

yrLDCT X-ray

Kanekoet al. (129) 1369 15 15.6 6.6 0.43 0.12 .20 .50Soneet al. (130) 3967 19 5.0 8.8 0.46–0.5 .30b 40–74Henschkeet al. (131) 1000 27 20.1 11.6 2.7 0.70 .10c .60

a Defined as individuals with “test-positive,” in whom further workup gave no suspicion of malignancy.b The study also included a group of nonsmokers.c Average5 45 (not reported in the other studies).

Table 5 Histology, stage, and size of primary lung cancer detected by low-dose spiral CT

AuthorNo. of cancers/No. screened

Histology % TNM % Size (mm)

Adenoa Squam. Other I II III IV Average Range ,10 11–20 .21

Kanekoet al. (129) 15/1369 (1.1%) 73 17 93 7 12 8–18Soneet al. (130) 19/5483 (0.3%) 63 5 32 84 16 17 6–47 4 14 3Henschkeet al. (131) 27/1000 (2.7%) 67 3 30 85 4 11 15 8 4

a Adeno, adenocarcinoma; Squam., squamous cell carcinoma; TNM, tumor-node-metastasis.




6–10-mm nodules (25). Given these promising preliminary clin-ical results, further research is needed to determine the optimaltechnique for spiral CT screening, which includes collimation,reconstruction interval, pitch, and viewing methods. Decreasingthe slice thickness to 3 mm, monitoring the viewing of exami-nations, and computer-aided diagnosis have been used to im-prove the diagnostic capability of spiral CT in the detection ofpulmonary nodules (133–136).

Future large scale randomized studies have to confirmwhether in fact spiral CT screening will lead to a reduction inlung cancer mortality. In a randomized study, the followingquestions arise: (a) what is the optimal high-risk group to studyand what should be the control arm? (b) what should be the endpoints (goals) of the studies? The ultimate goal is to reduce thelung cancer mortality. However, although this is a long-termgoal, intermediate end points from such studies should be eval-uated. The change to more curable stages at diagnosis for thelung cancer patients is one such immediate goal; (c) what is theoptimal workup and the morbidity of this program? (d) what isthe cost of such a screening program? and (e) what is thefalse-positive rate of the screening findings? Incorporation ofsmoking cessation programs should be included in the futuredesign of screening studies because it has been shown thatscreening with low-dose CT in participants who are still smok-ing provides substantial motivation for smoking cessation (137).

The studies with spiral CT-scan have demonstrated thesuperior diagnostic ability in the detection of small peripherallylocated tumors, most of the malignant ones of adenocarcinomatype of histology. The diagnostic sensitivity of spiral CT formore centrally located tumors (mostly squamous cell carci-noma) is significantly lower than for the peripherally locatedones. Through these spiral CT studies, we will learn about thebiology, pathology, and clinical course of these small tumors,which might be different from what we know about clinicallymore evident tumors detected routinely in previous studies.

Because lung cancer is so common, the introduction of anynew screening technique in this area has to be underpinned bycareful definition of the cost implications and must be justifiedby compelling evidence. The cost-effectiveness of the spiralCT approach should be assessed by evaluating the rate ofover-diagnosing nonmalignant, relatively common abnormali-ties and comparing CT imaging to other diagnostic technologies.

PET with FDG has recently emerged as a practical anduseful imaging modality in the preoperative staging of patientswith lung cancer. However, whereas CT is most frequently usedto provide additional anatomical and morphological informationabout lesions, the FDG PET imaging provides physiological andmetabolic information that characterizes lesions that are inde-terminate by CT. FDG PET imaging takes advantage of theincreased accumulation of FDG in transformed cells and issensitive (;95%) for the detection of cancer in patients whohave indeterminate lesions on CT (138). The specificity (;85%)of PET imaging is slightly less than its sensitivity because someinflammatory processes avidly accumulate FDG. The high neg-ative predictive value of PET suggests that lesions considerednegative on the study are benign, biopsy is not needed, andradiographic follow-up is recommended. Several studies havedocumented the increased accuracy of PET compared with CTin the evaluation of the hilar and mediastinal lymph node status

in patients with lung cancer (138). However, the PET resolutionis sufficient only for nodules$6 cm and will not be helpful indetecting the very small nodules. Compared with low-dosespiral CT, the FDG PET scan is more expensive and timeconsuming. The role of PET scan in early diagnosis of lungcancer in an asymptomatic high-risk population is not yet eval-uated. However, future studies have to include PET evaluationto define its role in a population screening setting.

ConclusionRecent advances in molecular biology and pathology have

led to a better understanding and documentation of morpholog-ical changes in the bronchial epithelium before development ofclinical evident lung carcinomas. Combined with technical de-velopments in radiological and bronchoscopic techniques, theseprocedures offer great promise in diagnosing lung cancer far inadvance of clinical presentation. Any of these individual proce-dures could be incorporated into the routine management ofindividuals at risk for developing primary or secondary lungcancer, and for several of these methods, clinical studies areunder way. Preliminary reported data are very promising for theearly detection of lung cancer. Future studies must incorporatethe different methods in a multidisciplinary scientific setting toevaluate the role of the individual method in the overall man-agement for individuals at high risk for developing lung cancer.Several of these tests might diagnose the disease at the stage ofclonal expansion before invasive carcinoma has developed. Amanagement and intervention strategy appropriate to that stageof disease have to be developed. Preliminary studies of chemo-prevention agents are reported, and new agents based on otherbiological mechanisms are under development and ready forclinical trials. It is now time to plan clinical trials that evaluateboth diagnostic and therapeutic approaches to access their im-pact on the incidence of clinical lung cancer.

AcknowledgmentsWe thank Drs. Stephen Lam, Vancouver, British Columbia, Can-

ada, and Kavita Garg, University of Colorado Health Sciences Center,Denver, Colorado, for a critical review of the manuscript and Drs.Timothy Kennedy and York Miller for submitting illustrations forwhite-light and LIFE bronchoscopy.

References1. Greenlee, R. T., Murray, T., Bolden, S., and Wingo, P. A. CancerStatistics, 2000. CA Cancer J. Clin.,50: 7–30, 2000.2. Mountain, C. T. Revision in the international system for staging oflung cancer. Chest,111: 1710–1717, 1997.3. Ihde, D. C. Chemotherapy of lung cancer. N. Engl. J. Med.,327:1434–1441, 1992.4. Melamed, M. R., Flehinger, B. J., Zaman, M. B., Heelan, R. T.,Parchick, W. A., and Martini, N. Screening for lung cancer: results ofthe Memorial Sloan-Kettering study in New York. Chest,86: 44–53,1984.5. Fontana, R. S., Sanderson, D. R., Woolner, L. B., Taylor, W. F.,Miller, W. E., and Muhm, J. R. Lung Cancer Screening. The Mayoprogram. J. Occup. Med.,28: 746–750, 1986.6. Tockman, M. S. Survival and mortality from lung cancer in ascreened population: The Johns Hopkins Study. Chest,89: 324S–325S,1986.




7. Kubik, A., Parkin, D. M., Khlat, M., Erban, J., Polak, J., andAdfamec, M. Lack of benefit from semi-annual screening for cancer ofthe lung: follow-up report of a randomized controlled trial on a popu-lation of high-risk males in Czechoslovakia. Int. J. Cancer,45: 26–33,1990.

8. Fontana, R. S., Sanderson, D. R., Woolner, L. B., Taylor, W. F.,Miller, W. E., Muhm, J. R., Bernatz, P. E., Payne, W. S., Pairolero,P. C., and Bergstralh, E. J. Screening of lung cancer. A critique of theMayo Lung Project. Cancer (Phila.),67: 1155–1164, 1991.

9. Strauss, G. M., Gleason, R. E., and Sugarbaker, D. J. Screening forlung cancer. Another look: a different view. Chest,111:754–768, 1997.

10. Hirsch, F. R., Brambilla, E., Gray, N., Gritz, E., Kelloff, G. J.,Linnoila, I., Pastorino, U., and Mulshine, J. L. Prevention and earlydetection of lung cancer—clinical aspects. Lung Cancer (Limerick),17:163–174, 1997.

11. Hong, W. K. Chemoprevention of Lung Cancer. Oncology (Basel),13 (Suppl. 5):135–141, 1999.

12. Mulshine, J. L., and Henschke, C. I. Prospects for lung-cancerscreening. Lancet,355: 592–593, 2000.

13. Travis, W. D., Colby, T. V., Corrin, B., Shimosato, Y., and Bram-billa, E. Histological typing of tumours of lung and pleura.In: L. H.Sobin (ed.), World Health Organization International Classification ofTumours, Ed. 3. New York: Springer-Verlag, 1999.

14. Franklin, W. A. Pathology of lung cancer. J. Thorac. Imaging,15:3–12, 2000.

15. Colby, T. V. Precursor lesions to pulmonary neoplasia.In: C.Brambilla and E. Brambilla (eds.), Lung Tumors. Fundamental Biologyand Clinical Management, pp. 61–87. New York: Marcel Dekker Inc.,1999.

16. Peters, E. J., Morice, R., Benner, S. E., Lippman, S., Lukeman, J.,Lee, J. S., Ro, J. L., and Hong, W. K. Squamous metaplasia of thebronchial mucosa and its relationship to smoking. Chest,103: 1429–1432, 1993.

17. Auerbach, O., Gere, B., Forman, J. B., Petrick, T. G., Smolin, H. J.,Muchsam, G. E., Kassouny, D. Y., and Stout, A. P. Changes in bron-chial epithelium in relation to smoking and cancer of the lung. N. Engl.J. Med.,256: 97–104, 1957.

18. Wistuba, I. I., Behrens C. Milchgrub, S., Bryant, D., Hung, J.,Minna, J. D., and Gazdar, A. F. Sequential molecular abnormalities areinvolved in the multistage development of squamous cell lung carci-noma. Oncogene,18: 643–650, 1999.

19. Lam, S., LeRiche, J. C., Zheng, Y., Coldman, A., MacAulay, C.,Hawk, E., Kelloff, G., and Gazdar, A. F. Sex-related differences inbronchial epithelial changes associated with tobacco smoking. J. Natl.Cancer Inst.,91: 691–696, 1999.

20. Saccomanno G. Archer, V. E., Auerbach, O., Saunders, R. P., andBrennan, L. M. Development of carcinoma of the lung as reflected inexfoliated cells. Cancer (Phila.),32: 256–270, 1974.

21. Frost, J. K., Ball, W, C., Jr., Levin, M. L., Tockman, M. S., Erozan,Y. S., Gupta, K., Eggleston, J. C., Pressman, N. J., Donithan, M. P., andKimball, A. W. Sputum cytopathology: use and potential in monitoringthe workplace environment by screening for biological effects of expo-sure. J. Occup. Med.,28: 692–703, 1986.

22. Venman, B. J. W., van Boxem, T. J. M., Smit, E. F., Postmus, P. E.,and Sutedja, T. G. Outcome of bronchial carcinomain situ. Chest,117:1572–1576, 2000.

23. Slebos, R. J., Baas, I. O., Clement, M. J., Offerhaus, G. J., Askin,F. B., Hruban, R. H., and Westra, W. H. p53 alterations in atypicalalveolar hyperplasia of the human lung. Hum. Pathol.,29: 801–808,1998.

24. Kitamura, H., Kameda, Y., Ito, T., and Hayashi, H. Atypical ade-nomatous hyperplasia of the lung. Implications for the pathogenesis ofperipheral lung adenocarcinoma. Am. J. Clin. Pathol.,111: 610–622,1999.

25. Diedrich, S., Semik, M., Lentschig, M. G., Winter, F., Scheld,H. H., Roos, N., and Bongartz, G. Helical CT of pulmonary nodules in

patients with extrathoracic malignancy: CT-surgical correlation. Am. J.Roentgenol.,172: 353–360, 1999.

26. Keith, R. L., Miller, Y. E., Gemmill, R. M., Drabkin, H. A.,Dempsey, E. C., Kennedy, T. C., Prindiville, S., and Franklin, W. A.Angiogenic squamous dysplasia in bronchi of individuals at high risk forlung cancer. Clin. Cancer Res.,6: 1616–1625, 2000.

27. Muller, K. M., and Muller, G. The ultrastructure of preneoplasticchanges in bronchial mucosa. Curr. Top. Pathol.,73: 233–263, 1983.

28. Hirsch, F. R., Gazdar, A. F., Gabrielson, E., Lam, S., and Franklin,W. A histopathologic evaluation of premalignant and early malignantbronchial lesions: an interactive program based on internet digital im-ages to improve WHO criteria for early diagnosis of lung cancer and formonitoring chemoprevention studies—a SPORE collaborative project.Lung Cancer (Limerick),29 (Suppl. 2):209, 2000.

29. Lengauer, C., Kinzler, K. W., and Vogelstein, B. Genetic instabil-ities in human cancers. Nature (Lond.),396: 643–649, 1998.

30. Fong, K. M., Sekido, Y., and Minna, J. D. Molecular pathogenesisof lung cancer. J. Thorac. Cardiovasc. Surg.,118: 1136–1152, 1999.

31. Hirano, T., Franzen, B., Kato, H., Ebihara, Y., and Auer, G. Genesisof squamous cell lung carcinoma: sequential changes of proliferation,DNA ploidy, and p53 expression. Am. J. Pathol.,144: 296–302, 1994.

32. Betticher, D. C., Heighway, J., Thatcher, N., and Hasleton, P. S.Abnormal expression of CCND1 and RB1 in resection margin epitheliaof lung cancer patients. Br. J. Cancer,75: 1761–1768, 1997.

33. Satoh, Y., Ishikawa, Y., Nakagawa, K., Hirano, T., and Tsuchiya, E.A follow-up study of progression from dysplasia to squamous cellcarcinoma with immunohistochemical examination of p53 protein over-expression in the bronchi of ex-chromate workers. Br. J. Cancer,75:678–683, 1997.

34. Li, Z. H., Zheng, J., Weiss, L. M., and Shibata D. c-kras, and p53mutations occur very early in adenocarcinoma of the lung. Am. J.Pathol.,144: 303–309, 1994.

35. Brambilla, E., Gazzeri, S., Lantuejoul, S., Coll, J. L., Moro, D.,Negoescu, A., and Brambilla, C. p53 mutant immunophenotype andderegulation of p53 transcription pathway (Bc12, Bax, and Waf1) inprecursor bronchial lesions of lung cancer. Clin. Cancer Res.,4: 1609–1618, 1998.36. Hung, J., Kishimoto, Y., Sugio, K., Virmani, A., McIntire, D. D.,Minna, J. D., and Gazdar, A. F. Allele-specific chromosome 3p dele-tions occur at an early stage in the pathogenesis of lung carcinoma.J. Am. Med. Assoc.,273: 558–563, 1995.37. Kishimoto, Y., Sugio, K., Hung, J. Y., Virmani, A. K., McIntire,D. D., Minna, J. D., and Gazdar, A. F. Allele-specific loss in chromo-some 9p loci in preneoplastic lesions accompanying non-small cell lungcancers. J. Natl. Cancer Inst.,87: 1224–1229, 1995.38. Sugio, K., Kishimoto, Y., Virmani, A. K., Hung, J. Y., and Gazdar,A. F. K-ras mutations are a relatively late event in the pathogenesis oflung carcinomas. Cancer Res.,54: 5811–5815, 1994.39. Wistuba, I. I., Behrens, C., Virmani, A. K., Milchgrub, S., Syed, S.,Lam, S., Mackay, B., Minna, J. D., and Gazdar, A. F. Allelic losses atchromosome 8p21–23 are early and frequent events in the pathogenesisof lung cancer. Cancer Res.,59: 1973–1979, 1999.40. Varella-Garcia, M., Gemmill, R. M., Rabenhorst, S. H., Lotto, A.,Drabkin, H. A., Archer, P. A., and Franklin, W. A. Chromosomalduplication accompanies allelic loss in non-small cell lung carcinoma.Cancer Res.,58: 4701–4707, 1998.41. Westra, W. H., Baas, I. O., Hruban, R. H., Askin, F. B., Wilson, K.,Offerhaus, G. J., Slebos, R. J. K-ras oncogene activation in atypicalalveolar hyperplasias of the human lung. Cancer Res.,56: 2224–2228,1996.42. Wistuba, I. I., Lam, S., Behrens, C., Virmani, A. K., Fong, K. M.,LeRiche, J., Samet, J. M., Srivastava, S., Minna, J. D., and Gazdar, A. F.Molecular damage in the bronchial epithelium of current and formersmokers. J. Natl. Cancer Inst.,89: 1366–1377, 1997.43. Mao, L., Lee, J. S., Kurie, J. M., Fan, Y. H., Lippman, S. M., Lee,J. J., Ro, J. Y., Broxson, A., Yu, R., Morice, R. C., Kemp, B. L., Khuri,F. R., Walsh, G. L., Hittelman, W. N., and Hong, W. K. Clonal genetic




alterations in the lungs of current and former smokers. J. Natl. CancerInst., 89: 857–862, 1997.

44. Smith, A. L., Hung, J., Walker, L., Rogers, T. E., Vuitch, F., Lee,E., and Gazdar, A. F. Extensive areas of aneuploidy are present in therespiratory epithelium of lung cancer patients. Br. J. Cancer,73: 203–209, 1996.

45. Franklin, W. A., Gazdar, A. F., Haney, J., Wistuba, I. I., La Rosa,F. G., Kennedy, T., Ritchey, D. M., and Miller, Y. E. Widely dispersedp53 mutation in respiratory epithelium: a novel mechanism for fieldcarcinogenesis. J. Clin. Investig.,100: 2133–2137, 1997.

46. Johnston, W. W., and Elson, C. E. Respiratory tract.In: M. Bibbo(ed.), Comprehensive Cytopathology, pp. 325–401. Philadelphia: Saun-ders, 1997.

47. Travis, W. D., Travis, L. B., and Devesa, S. S. Lung cancer. Cancer(Phila.),75 (Suppl. 1):191–202, 1995.

48. Wynder, E. L., and Muscat, J. E. The changing epidemiology ofsmoking and lung cancer histology. Environ. Health Perspect.,103(Suppl. 8):143–148, 1995.

49. Thun, M. J., Lally, C. A., Flannery, J. T., Calle, E. E., Flanders,W. D., and Heath, C. W., Jr. Cigarette smoking and changes in thehistopathology of lung cancer. J. Natl. Cancer Inst.,89: 1580–1586,1997.

50. Lam, S., and Shibuya, H. Early diagnosis of lung cancer. Clin.Chest Med.,20: 53–61, 1999.

51. Kennedy, T. C., Proudfoot, S., Piantadosi, S., Wu, L., Saccomanno,G., Petty, T. L., and Tockman, M. S. Efficacy of two sputum collectiontechniques in patients with air flow obstruction. Acta Cytol.,43: 630–636, 1999.52. Browman, G. P., Arnold, A., Levine, M. N., and D’Souza, T. Useof screening phase data to evaluate observer variation of sputum cyto-diagnosis as an outcome measure in a chemoprevention trial. CancerRes.,50: 1216–1219, 1990.53. Cantaboni, A., Pezzotta, M. G., Sironi, M., and Porcellati, M.Quality assurance in pathology: cytologic and histologic correlation.Acta Cytol.,36: 717–721, 1992.54. Holiday, D. B., McLarty, J. W., Farley, M. L., Mabry, L. C.,Cozens, D., Roby, T., Waldron, E., Underwood, R. D., Anderson, E.,and Culbreth, W. Sputum cytology within and across laboratories. Areliability study. Acta Cytol.,39: 195–206, 1995.55. Risse, E. K., Vooijs, G. P., and van’t Hoff, M. A. Relationshipbetween the cellular composition of sputum and the cytologic diagnosisof lung cancer. Acta Cytol.,31: 170–176, 1987.56. MacDougall, B., and Weinerman, B. The value of sputum cytology.J. Gen. Intern. Med.,7: 11–12, 1992.57. Tockman, M. S., Erozan, Y. S., Gupta, P., Piantadosi, S., Mulshine,J. L., and Ruckdeschel, J. C. The early detection of second primary lungcancers by sputum immunostaining. Chest,106 (Suppl.): 385S–390S,1994.58. Tockman, M. S., Gupta, P. K., Myers, J. D., Frost, J. K., Baylin,S. B., Gold, E. B., Chase, A. M., Wilkinson, P. H., and Mulshine, J. L.Sensitive and specific monoclonal antibody recognition of human lungcancer antigen on preserved sputum cells: a new approach to early lungcancer detection. J. Clin. Oncol.,6: 1685–1693, 1988.59. Zhou, J., Mulshine, J. L., Unsworth, E. J., Scott, F. M., Avis, I. M.,Vos, M. D., and Treston, A. M. Purification and characterization of aprotein that permits early detection of lung cancer. J. Biol. Chem.,271:10760–10766, 1996.60. Qiao Y-L, Tockman, M. S., Li, L., Erozan, Y. S., Yao, S. X.,Barrett, M. J., Zhou, W. H., Giffen, C. A., Luo, X. C., and Taylor, P. R.A case-cohort study of an early biomarker of lung cancer in a screeningcohort of Yunnan tin miners in China. Cancer Epidemiol. Biomark.Prev,6: 893–900, 1997.61. Grover, F. L., and Piantadosi, S. Recurrence and survival followingresection of bronchioloalveolar carcinoma of the lung: the Lung CancerStudy Group experience. Ann. Surg.,209: 779–790, 1989.62. Tockman, M. S. Advances in sputum analysis for screening andearly detection of lung cancer. Cancer Control,7: 19–24, 2000.

63. Sueoka, E., Goto, Y., Sueoka, N., Kai, Y., Kozu, T., and Fujiki, H.Heterogeneous nuclear ribonucleoprotein B1 as a new marker of earlydetection for human lung cancers. Cancer Res.,59: 1404–1407, 1999.

64. Mulshine, J. L. Reducing lung cancer risk. Early detection. Chest,116: 493S–496S, 1999.

65. Mao, L., Hruban, R. H., Boyle, J. O., Tockman, M., and Sidransky,D. Detection of oncogene mutations in sputum precedes diagnosis oflung cancer. Cancer Res.,54: 1634–1637, 1994.

66. Ruppert, J. M., Tokino, K., and Sidransky, D. Evidence for twobladder cancer suppressor loci on chromosome 9. Cancer Res.,53:5093–5095, 1993.

67. Nawroz, H., van der Riet, P., Hruban, R. H., Koch, W., Ruppert,J. M., and Sidransky, D. Allelotype of head and neck squamous cellcarcinoma. Cancer Res.,54: 1152–1155, 1994.

68. Miozzo, M., Sozzi, G., Musso, K., Pilotti, S., Incarbone, M., Pas-torino, U., and Pierotti, M. A. Microsatellite alterations in bronchial andsputum specimens of lung cancer patients. Cancer Res.,56: 2285–2288,1996.

69. Shapiro, G. I., Park, J. E., Edwards, C. D., Mao, L., Merlo, A.,Sidransky, D., Ewen, M. E., and Rollins, B. J. Multiple mechanisms ofp16INK4A inactivation in non-small cell lung cancer cell lines. CancerRes.,55: 6200–6209, 1995.

70. Hamada, K., Kohno, T., Kawanishi, M., Ohwada, S., and Yokota, J.Association of CDKN2A(p16)/CDKN2B9p15) alterations and homozy-gous chromosome arm 9p deletions in human lung carcinoma. GenesChromosomes Cancer,22: 232–240, 1998.

71. Belinsky, S. A., Nikula, K. J., Palmisano, W. A., Michels, R.,Saccomanno, G., Gabrielson, E., Baylin, S. B., and Herman, J. G.Aberrant methylation of p16ink4a is an early event in lung cancer and apotential biomarker for early diagnosis. Proc. Natl. Acad. Sci. USA,95:11891–11896, 1998.

72. Esteller, M., Sanchez-Cespedes, M., Rosell, R., Sidransky, D.,Baylin, S. B., and Herman, J. G. Detection of aberrant promoter hyper-methylation of tumor suppressor genes in serum DNA from non-smallcell lung cancer patients. Cancer Res.,59: 67–70, 1999.73. Nieburgs, H. E. Recent progress in the interpretation of malignancyassociated changes (MAC). Acta Cytol.,12: 445–453, 1968.74. Payne, P. W., Sebo, T. J., Doudkine, A., Garner, D., MacAulay, C.,Lam, S., LeRiche, J. C., and Palcic, B. Sputum screening by quantitativemicroscopy: a reexamination of a portion of the National Cancer Insti-tute Cooperative Early Lung Cancer Study. Mayo Clin. Proc.,72:697–704, 1997.75. Lam, S., Palcic, B., Garner, D., Beveridge, J., MacAulay, C.,LeRiche, J., and Coldman, A. Lung Cancer Control Strategy in the NewMillennium. Lung Cancer,29 (Suppl. 2):145, 2000.76. Ahrendt, S. A., Chow, J. T., Xu, L. H., Yang, S. C., Eisenberger,C. F., Esteller, M., Herman, J. G., Wu, L., Decker, P. A., Jen, J., andSidransky, D. Molecular detection of tumor cells in bronchoalveolarlavage fluid from patients with early stage lung cancer. J. Natl. CancerInst., 91: 332–339, 1999.77. Gazdar, A. F., and Minna, J. D. Molecular detection of early lungcancer. J. Natl. Cancer Inst.,91: 299–301, 1999.78. Field, J. K., Liloglou, T., Xinarianos, G., Prime, W., Fielding, P.,Walshaw, M. J., and Turnbull, L. Genetic alterations in bronchial lavageas a potential marker for individuals with a high risk of developing lungcancer. Cancer Res.,59: 2690–2695, 1999.79. Liloglou, T., Maloney, P., Xinarianos, G., Fear, S., and Field, J. K.Sensitivity and limitations of high throughput fluorescent microsatelliteanalysis for the detection of allelic imbalance. Application in lungtumors. Int. J. Oncol,16: 5–14, 2000.80. Fielding, P., Turnbull, L., Prime, W., Walshaw, M., and Field, J. K.Heterogeneous nuclear ribonucleoprotein A2/B1 up-regulation in bron-chial lavage specimens: a clinical marker of early lung cancer detection.Clin. Cancer Res.,5: 4048–4052, 1999.81. Pantel, K., Schlimok, G., Angstwurm, M., Weckermann, D.,Schmaus, W., Gath, H., Passlick, B., Izbicki, J. R., and Riethmuller, G.Methodological analysis of immunocytochemical screening for dissem-




inated epithelial tumor cells in bone marrow. J. Hematother.,3: 165–173, 1994.

82. Lambrechts, A. C., van’t Veer, L. J., and Rodenhuis, S. The detec-tion of minimal numbers of contaminating epithelial tumor cells inblood or bone marrow: use, limitations and future of RNA-based meth-ods. Ann. Oncol.,9: 1269–1276, 1998.

83. Steinman, C. R. Free DNA in serum and plasma from normaladults. J. Clin. Investig.,66: 1391–1399, 1980.

84. Raptis, L., and Menard, H. A. Quantitation and characterization ofplasma DNA in normals and patients with lupus erythematosus. J. Clin.Investig.,66: 1391–1399, 1980.

85. Leon, S. A., Shapiro, B., Sklaroff, D. M., and Yaros, M. J. FreeDNA in the serum of cancer patients and the effect of therapy. CancerRes.,37: 646–650, 1977.

86. Stroun, M., Anker, P., Maurice, P., Lyautey, J., Lederrey, C., andBeljanski, M. Neoplastic characteristics of the DNA found in the plasmaof cancer patients. Oncology,46: 318–322, 1989.

87. Shapiro, B., Chakrabaty, M., Cohn, E., and Leon, S. A. Determi-nation of circulating DNA levels in patients with benign or malignantgastrointestinal disease. Cancer (Phila.),51: 2116–2120, 1983.

88. Chen, X. Q., Stroun, M., Magnenat, J-L., Nicod, L. P., Kurt, A. M.,Lyautey, J., Lederrey, C., and Anker, P. Microsatellite alterations inplasma DNA of small cell lung cancer patients. Nat. Med.,2: 1033–1037, 1996.

89. Anker, P., Lefort, F., Vasioukhin, V., Lyautey, J., Lederrey, C.,Chen, X. Q., Stroun, M., Mulcahy, H. E., and Farthing, M. J. K-rasmutations are found in DNA extracted from the plasma of colorectalcancer patients. Gastroenterology,112: 1114–1129, 1997.

90. Kopreski, M. S., Benko, F. A., Kwee, C., Leitzel, K. E., Eskander,E., Lipton, A., and Gocke, C. D. Detection of mutant K-ras DNA inplasma or serum of patients with colorectal cancer. Br. J. Cancer,76:1293–1299, 1997.

91. Hibi, K., Robinson, R., Wu, L., Hamilton, S. R., Sidransky, D., andJen, J. Molecular detection of genetic alterations in the serum of colo-rectal cancer patients. Cancer Res.,58: 1405–1407, 1998.

92. Sorenson, G. D., Pribish, D. M., Valone, F. H., Memoli, V. A.,Bzik, D. J., and Yao, S. L. Soluble normal and mutated DNA sequencesfrom single copy genes in human blood. Cancer Epidemiol. Biomark.Prev.,3: 67–71, 1994.

93. Mulcahy, H. E., Lyautey, J., Lederrey, C., qi Chen, X., Anker, P.,Alstead, E. M., Ballinger, A., Farthing, M. J., and Stroun, M. A pro-spective study of K-rasmutations in the plasma of pancreatic cancerpatients. Clin. Cancer Res.,4: 271–275, 1998.

94. Vasioukhin, V., Anker, P., Maurice, P., Lyautey, J., Lederrey, C.,and Stroun, M. Point mutations of the N-ras in the blood plasma DNAof patients with myelodysplastic syndrome or acute myelogenous leu-kaemia. Br. J. Haematol.,86: 774–779, 1994.

95. Woolner, L. B. Pathology of cancer detected cytologically.In: Atlasof Early Lung Cancer, National Cancer Institute, NIH, United StatesDepartment of Health and Human Services, pp. 107–203. Tokyo: Igaku-Shoin, 1983.

96. Kato, H., and Cortese, D. A. Early detection of lung cancer bymeans of hematoporphyrin derivative fluorescence and laser photo-radiation. Clin. Chest Med.,6: 237–253, 1985.

97. Profio, A. E., Doiron, D. R., and King, E. G. L. Laser fluorescencebronchoscopy for localization of occult lung tumors. Med. Phys. (NY),6: 532–535, 1979.

98. Kinsey, J. H., and Cortese, D. A. Endoscopic system for simulta-neous visual examination and electronic detection of fluorescence. Rev.Sci. Instrum.,51: 1403–1406, 1980.

99. Profio, A. E., Doiron, D. R., and Sarnaik, J. Fluorometer forendoscopic diagnosis of tumors. Med. Phys. (NY),11: 516–520, 1984.100. Montan, S., Svanberg, K., and Svanberg, S. Multicolor imagingand contrast enhancement in cancer-tumor localization using laser-induced fluorescence in hematoporphyrin-derivative-bearing tissue. Op-tics Lett.,10: 56–58, 1985.

101. Kato, H., Imaizumi, T., Aizawa, K., Iwabuchi, H., Yamamoto, H.,Ikeda, N., Tsuchida, T., Tamachi, Y., Ito, T., and Hayata, Y. Photody-namic diagnosis in respiratory tract malignancy, using an excimer dyelaser system. J. Photochem. Photobiol. Biol.,6: 189–196, 1990.

102. Kato, H., and Ikeda, N. The role of fluorescence diagnosis in theearly detection of high-risk bronchial lesions. J. Bronchol.,5: 273–274,1998.

103. Palcic, B., Lam, S., Hung, J., and MacAulay, C. Detection andlocalization of early lung cancer by imaging techniques. Chest,99:742–743, 1991.

104. Hung, J., Lam, S., LeRiche, J. C., and Palcic, B. Autofluorescenceof normal and malignant bronchial tissue. Lasers Surg. Med.,11:99–105, 1991.

105. Lam, S., Kennedy, T., Unger, M., Miller, Y. E., Gelmont, D.,Rusch, V., Gipe, B., Howard, D., LeRiche, J. C., Coldman, A., andGazdar, A. F. Localization of bronchial intraepithelial neoplastic lesionsby fluorescence bronchoscopy. Chest,113: 696–702, 1998.

106. Kurie, J. M., Lee, J. S., Morice, R. C., Walsh, G. L., Khuri, F. R.,Broxson, A., Ro JY. Franklin, W. A., Yu, R., and Hong, W. K.Autofluorescence bronchoscopy in the detection of squamous metapla-sia and dysplasia in current and former smokers. J. Natl. Cancer Inst.,90: 991–995, 1998.

107. Venmans, B. J., van der Linden, J. C., van Boxem, A. J., Postmus,P. E., Smit, E. F., and Sutedja, G. Early detection of pre-invasive lesionsin high risk patients. A comparison of conventional fiberoptic andfluorescence bronchoscopy. J. Bronchol.,5: 280–283, 1998.108. Vermylen, P., Pierard, P., Roufosse, C., Bosschaerts, T., Verhest,A., Sculier, J-P, and Ninane, V. Detection of bronchial preneoplasticlesions and early lung cancer with fluorescence bronchoscopy: a studyabout its ambulatory feasibility under local anaesthesia. Lung Cancer(Limerick), 25: 161–168, 1999.109. Kennedy, T., Hirsch, F. R., Miller, Y., Prindiville, S., Bunn, P. A.,Jr., and Franklin, W. A randomized study of fluorescence bronchoscopyversus white-light bronchoscopy for early detection of lung cancer inhigh risk patients. Lung Cancer (Limerick),29 (Suppl. 2):244, 2000.110. Gazdar, A. F., and Minna, J. D. Angiogenesis and the multistagedevelopment of lung cancers. Clin. Cancer Res.,6: 1611–1612, 2000.111. Henschke, C. I., Miettinen, O. S., Yankelevitz, D. F., Libby,D. M., and Smith, J. P. Radiographic screening for cancer: proposedparadigm for requisite research. Clin. Imaging,18: 6–20, 1994.112. Miettinen, O. S. Screening for lung cancer. Radiol. Clin. N. Am.,38: 479–486, 2000.113. Marcus, P. M., Bergstralh, E. J., Fagerstrom, M., Willliams, D. E.,Fontana, R., Taylor, W. F., and Prorok, P. C. Lung cancer mortality inthe Mayo Lung Project: impact of extended follow-up. J. Natl. CancerInst., 92: 1308–1316, 2000.114. Black, W. C. Overdiagnosis. An underrecognized cause of confu-sion and harm in cancer screening. J. Natl. Cancer Inst.,92: 1280–1282,2000.115. The Health, and Welfare Statistics Foundation. The current issueon the screening project for elderly people by the Ministry of Health andWelfare Japan. J. Health Welfare,42: 37, 1995.116. Sobue, T., Suzuki, T., and Naruke, T. The Japanese Lung-Cancer-Screening Research Group; a case-control study for evaluating lungcancer screening in Japan. Int. J. Cancer,50: 230–237, 1992.117. Okamoto, N., Suzuki, T., Hasegawa, H., Gotoh, T., Hagiware, S.,Sekimoto, M., and Kaneko, M. Evaluation of a clinic-based screeningprogram for lung cancer with a case-control design in Kanagawa. Japan.Lung Cancer (Limerick),25: 77–85, 1999.118. Muhm, J. R., Miller, W. F., Fontana, R. S., Sanderson, D. R., andUhlenhopp, M. A. Lung cancer detected during a screening programusing four-month chest radiographs. Radiology,148: 609–615, 1983.119. Hayabuchi, N., Russell, W. J., and Murakami, J. Problems inradiographic detection and diagnosis of lung cancer. Acta Radiol.(CPH),30: 163–167, 1989.120. Woodring, J. H. Pitfalls in the radiologic diagnosis of lung cancer.Am. J. Roentgenol.,154: 1165–1175, 1990.




121. Greene, R. E. Missed lung nodules: lost opportunities for cancercare. Radiology,182: 8–9, 1992.

122. Austin, J. H. M., Romney, B. M., Goldsmith, L. S. Missed bron-chogenic carcinoma: radiographic findings in 27 patients with a poten-tially resectable lesions evident in retrospect. Radiology,182: 115–122,1992.

123. Schaner, E. G., Chang, A. E., Doppan, J. L., Conkle, D. M., Flye,M. W., and Rosenberg, S. A. Comparison of computed and conventionalwhole lung tomography in detecting pulmonary nodules. Am. J. Roent-genol.,131: 51–54, 1978.

124. Webb, W. R. Radiologic evaluation of solitary pulmonary nodule.Am. J. Roentgenol.,154: 701–708, 1990.

125. Kalender, W. A., Seissler, W., Klotz, E., and Vock, P. Spiralvolumetric CT with single-breath-hold technique, continuous transport,and continuous scanner rotation. Radiology,176: 181–183, 1990.

126. Costello, P., Anderson, W. A., and Blume, D. Pulmonary nodule:evaluation with spiral volumetric CT. Radiology,179: 875–876, 1991.

127. Remy-Jardin, M., Giraud, F., and Marquette, C. H. Pulmonarynodules: detection with thick section spiral CT versus conventional CT.Radiology,187: 513–520, 1993.

128. Davis, S. CT evaluation for pulmonary metastases in patients withextrathoracic malignancy. Radiology,180: 1–12, 1991.

129. Kaneko, M., Eguchi, K., Ohmatsu, H., Kakinuma, R., Naruke, T.,Suemasu, K., and Moriyama, N. Peripheral lung cancer: screening anddetection with low-dose spiral CT versus radiography. Radiology,201:798–802, 1996.

130. Sone, S., Takashima, S., Li, F., Yang, Z., Honda, T., Maruyama,Y., Hasegawa, M., Yamanda, T., Kubo, K., Hanamura, K., and Asakura,K. Mass screening for lung cancer with mobile spiral computed tomog-raphy scanner. Lancet,351: 1242–1245, 1998.

131. Henschke, C. I., McCauley, D. I., Yankelevitz, D. F., Naidich,D. P., McGuinness, G., Miettinen, O. S., Libby, D. M., Pasmantier,M. W., Koizumi, J., Altorki, N. K., and Smith, J. P. Early Lung CancerAction Project: overall design and findings from baseline screening.Lancet,354: 99–105, 1999.

132. Kakinuma, R., Ohmatsu, H., Kaneko, M., Eguchi, K., Naruke, T.,Nagai, K., Nishiwaki, Y., Suzuki, A., and Moriyama, N. Detectionfailures in spiral CT screening for lung cancer: analysis of CT findings.Radiology,212: 61–66, 1999.

133. Paranjpe, D. V., and Bergin, C. J. Spiral CT of the lungs: optimaltechnique and resolution compared with conventional CT. Am. J. Roent-genol.,162: 561–567, 1994.

134. Seltzer, S. E., Judy, P. F., Adams, D. F., Jacobson, F. L., Stark, P.,Kikinis, R., Swensson, R. G., Hooton, S., Head, B., and Feldman, U.Spiral CT of the chest: comparison of cine and film-based viewing.Radiology,197: 73–78, 1995.

135. Croisille, P., Souto, M., Cova, M., Wood, S., Afework, Y., Kuhl-man, J. E., and Zerhouni, E. A. Pulmonary nodules: improved detectionwith vascular segmentation and extraction with spiral CT. Radiology,197: 397–401, 1995.

136. Kanazawa, K., Kawata, Y., Niki, N., Satoh, H., Ohmatsu, H.,Kakinuma, R., and Kaneko, M. Computer-aided diagnosis for pulmo-nary nodules based on helical CT images. Comput. Med. ImagingGraph.,22: 157–167, 1998.

137. Buckshee, N., Agnello, K., Yankelevitz, D. F., Mancuso, C., andHenschke, C. I. Smoking habits and overall satisfaction after early lungcancer screening using low-dose CT. ALA/ATS International Confer-ence, 1999.

138. Coleman, R. E. PET in lung cancer. J. Nucl. Med.,40: 814–820,1999.

139. Maitra A. Wisstuba, I. I., Virmani, A. K., Sakaguchi, M., Park, I.,Stucky, A., Milchgrub, S., Gibbons, D., Minna, J. D., and Gazdar, A. F.Enrichment of epithelial cells for molecular studies. Nat. Med.,5:459–463, 1999.140. Kopelovich, L., Henson, D. E., Gazdar, A. F., Dubb, B., Srivas-tava, S., Kelloff, G. J., and Greenwald, P. Surrogate anatomic/functional

sites for evaluating cancer risk: an extension of the field effect. Clin.Cancer Res., 5: 3899–3905, 1999.

141. Crowell, R. E., Gilliland, F. D., Temes, R. T., Harms, H. J., Neft,R. E., Heaphy, E., Auckley, D. H., Crooks, L. A., Jordan, S. W., Samet,J. M., Lechner, J. F., and Belinsky, S. A. Detection of trisomy 7 innonmalignant bronchial epithelium from lung cancer patients and indi-viduals at risk for lung cancer. Cancer Epidemiol. Biomark. Prev.,5:631–637, 1996.

142. Somers, V. A., van Henten, A. M., ten Velde, G. P., Arends, J. W.,and Thunnissen, F. B. Additional value of K-ras point mutations inbronchial wash fluids for diagnosis of peripheral lung tumours. Eur.Respir. J.,13: 1120–1124, 1999.

143. Yahata, N., Ohyashiki, K., Ohyashiki, J. H., Iwama, H., Hayashi,S., Ando, K., Hirano, T., Tsuchida, T., Kato, H., Shay, J. W., andToyama, K. Telomerase activity in lung cancer cells obtained frombronchial washings. J. Natl. Cancer Inst.,90: 684–690, 1998.

144. Wistuba, I. I., and Gazdar, A. F. Molecular abnormalities in thesequential development of lung carcinoma.In: S. Srivastava, D. E.Henson, and A. F. Gazdar (eds.), Molecular Pathology of Early Cancer,pp. 265–276. Amsterdam: IOS Press, 1999.

145. Yashima, K., Litzky, L. A., Kaiser, L., Rogers, T., Lam, S.,Wistuba, I. I., Milchgrub, S., Srivastava, S., Piatyszek, M. A., Shay,J. W., and Gazdar, A. F. Telomerase expression in respiratory epithe-lium during the multistage pathogenesis of lung carcinomas. CancerRes.,57: 2373–2377, 1997.

146. Scott, F. M., Treston, A. M., Shaw, G. L., Avis, I., Sorenson, J.,Kelly, K., Dempsey, E. C., Cantor, A. B., Tockman, M., and Mulshine,J. L. Peptide amidating activity in human bronchoalveolar lavage fluid.Lung Cancer (Limerick),14: 239–251, 1996.

147. Scott, F. M., Modali, R., Lehman, T. A., Seddon, M., Kelly, K.,Dempsey, E. C., Wilson, V., Tockman, M. S., and Mulshine, J. L. Highfrequency of K-ras codon 12 mutations in bronchoalveolar lavage fluidof patients at high risk for second primary lung cancer. Clin. CancerRes.,3: 479–482, 1997.

148. Mills, N. E., Fishman, C. L., Rom, W. N., and Jacobson, D. R.Rasoncogene detection in bronchioloalveolar lavage fluid from patientswith lung cancer. Lung Cancer (Limerick),1 (Suppl.):11, 1994.

149. Nawroz, H., Koch, W., Anker, P., Stroun, M., and Sidransky, D.Microsatellite alterations in serum DNA of head and neck cancer pa-tients. Nat. Med.,2: 1035–1037, 1996.

150. Brennan, J. A., Mao, L., Hruban, R. H., Boyle, J. O., Eby, Y. J.,Koch, W. M., Goodman, S. N., and Sidransky, D. Molecular assessmentof histopathological staging in squamous-cell carcinoma of the head andneck. N. Engl. J. Med.,332: 429–435, 1995.

151. Ahrendt, S. A., Yang, S. C., Wu, L., Westra, W. H., Jen, J.,Califano, J. A., and Sidransky, D. Comparison of oncogene mutationdetection and telomerase activity for the molecular staging of non-smallcell lung cancer. Clin. Cancer Res.,3: 1207–1214, 1997.

152. Gazdar, A. F., and Minna, J. D. NCI series of cell lines: anhistorical perspective. J. Cell. Biochem.,24 (Suppl.):1–11, 1996.153. Wistuba, I. I., Bryant, D., Behrens, C., Milchgrub, S., Virmani,A. K., Ashfaq, R., Minna, J. D., and Gazdar, A. F. Comparison offeatures of human lung cancer cell lines and their corresponding tumors.Clin. Cancer Res.,5: 991–1000, 1999.154. Amstad, P., Reddel, R. R., Pfeifer, A., Malan, S. L., Mark, G. D.,and Harris, C. C. Neoplastic transformation of a human bronchialepithelial cell line by a recombinant retrovirus encoding viral Harveyras. Mol. Carcinog.,1: 151–160, 1988.155. Franklin, W. A., Folkvord, J. M., Varella-Garcia, M., Kennedy, T.,Proudfoot, S., Cook, R., Dempsey, E. C., Helm, K., Bunn, P. A., andMiller, Y. E. Expansion of bronchial epithelial cell populations by invitro culture of explants from dysplastic and histologically normal sites.Am. J. Respir. Cell Mol. Biol.,15: 297–304, 1996.156. Kelsey, K., Spitz, M., Zuo, A., and Wiencke, J. Deletion ofglutathioneS-transferase classm and classu genes interacts to enhancesusceptibility to lung cancer in minority populations. Cancer CausesControl, 1997.




157. Wu, X., Zhao, Y., Honn, S. E., Tomlinson, G. E., Minna, J. D.,Hong, W. K., and Spitz, M. R. Benzo[a]pyrene diol epoxide-induced3p21.3 aberrations and genetic predisposition to lung cancer. CancerRes.,58: 1605–1608, 1998.158. Virmani, A. K., Fong, K. M., Kodagoda, D., McIntire, D., Hung,J., Tonk, V., Minna, J. D., and Gazdar, A. F. Allelotyping demonstratescommon and distinct patterns of chromosomal loss in human lungcancer types. Genes Chromosomes Cancer,21: 308–319, 1998.159. Takahashi, T., Nau, M. M., Chiba, I., Birrer, M. J., Rosenberg,R. K., Vinocour, M., Levitt, M., Pass, H., Gazdar, A. F., and Minna,J. D. p53: a frequent target for genetic abnormalities in lung cancer.Science (Washington DC),246: 491–494, 1989.160. Forgacs, E., Biesterveld, E. J., Sekido, Y., Fong, K. M., Muneer,S., Wistuba, I., Milchgrub, S., Brezinschek, R., Virmani, A., Gazdar,A. F., and Minna, J. D. Mutation analysis of thePTEN/MMAC1gene inlung cancer. Oncogene,17: 1557–1565, 1998.161. Mitsudomi, T., Steinberg, S., Oie, H. K., Mulshine, J. L., Phelps,R., Viallet, J., Pass, H., Minna, J. D., and Gazdar, A. F.ras genemutations in non-small cell lung cancers are associated with shortenedsurvival irrespective of treatment intent. Cancer Res.,51: 4999–5002,1991.162. Sozzi, G., Veronese, M. L., Negrini, M., Baffa, R., Corticelli,M. G., Inoue, H., Tornielli, S., Pilotti, S., Ohta, M., Huebner, K., andCroce, C. M. TheFHIT gene at 3p14.2 is abnormal in lung cancer. Cell,85: 17–26, 1996.163. Anbazhagan, R., Tihan, T., Bornman, D. M., Johnston, J. C., Saltz,J. H., Weigering, A., Piantadosi, S., and Gabrielson, E. Classification ofsmall cell lung cancer and pulmonary carcinoid by gene expressionprofiles. Cancer Res.,59: 5119–5122, 1999.164. Hibi, K., Liu, Q., Beaudry, G. A., Madden, S. L., Westra, W. H.,Wehage, S. L., Yang, S. C., Heitmiller, R. F., Bertelsen, A. H., Sidran-sky, D., and Jen, J. Serial analysis of gene expression in non-small celllung cancer. Cancer Res.,58: 5690–5694, 1998.165. Sozzi, G., Pastorino, U., Moiraghi, L., Tagliabue, E., Pezzella, F.,Ghirelli, C., Tornielli, S., Sard, L., Huebner, K., Pierotti, M. A., Croce,C. M., and Pilotti, S. Loss of FHIT function in lung cancer andpreinvasive bronchial lesions. Cancer Res.,58: 5032–5037, 1998.

166. Tockman, M. S., and Mulshine, J. L. Sputum screening by quan-titative microscopy: a new dawn for detection of lung cancer? MayoClin. Proc., 72: 788–790, 1997.167. Izzo, J., and Hittelman, W. N. Characterization of multistep tu-morigenesis byin situ hybridization. In: M. Andreeff and D. Pinkel(eds.), Introduction to FISH. In press, 2001.168. Neft, R. E., Crowell, R. E, Gilliland, F. D., Murphy, M. M., Lane,J. L., Harms, H., Coons, T., Heaphy, E., Belinsky, S. A., and Lechner,J. F. Frequency of trisomy 20 in nonmalignant bronchial epitheliumfrom lung cancer patients and cancer-free former uranium miners andsmokers. Cancer Epidemiol. Biomark. Prev,7: 1051–1054, 1998.169. Heppell-Parton, A. C., Nacheva, E., Carter, N. P., and Rabbitts,P. H. A combined approach of conventional and molecular cytogeneticsfor detailed karyotypic analysis of the small cell lung carcinoma cell lineU2020. Cancer Genet. Cytogenet.,108: 110–119, 1999.170. Lechner, J. F., Neft, R., Gilliland, F. D., Crowell, R. E., Auckely,D. H., Temes, R. T., and Belinsky, S. A. Individuals at high risk for lungcancer have airway epithelial cells with chromosome aberrations fre-quently found in lung tumor cells. In Vivo12: 23–26, 1998.171. Walch, A. K., Zitzelsberger, H. F., Aubele, M. M., Mattis, A. E.,Bauchinger, M., Candidus, S., Prauer, H. W., Werner, M., and Hofler,H. Typical and atypical carcinoid tumors of the lung are characterizedby 11q deletions as detected by comparative genomic hybridization.Am. J. Pathol.,153: 1089–1098, 1998.172. Petersen, I., Bujard, M., Petersen, S., Wolf, G., Goeze, A., Sch-wendel, A., Langreck, H., Gellert, K., Reichel, M., Just, K., du Manoir,S., Cremer, T., Dietel, M., and Ried, T. Patterns of chromosomalimbalances in adenocarcinoma and squamous cell carcinoma of thelung. Cancer Res.,57: 2331–2335, 1997.173. Smith, A. L., Hung, J., Walker, L., Rogers, T. E., Vuitch, F., Lee,E., and Gazdar, A. F. Extensive areas of aneuploidy are present in therespiratory epithelium of lung cancer patients. Br. J. Cancer,73: 203–209, 1996.174. MacAulay, C. E., Lam, S., Kein-Parker, H., Gazdar, A., Guillaud,M., Payne, P., LeRiche, J., Dawe, C., Band, P., and Palcic, B. Interme-diate endpoint biomarkers for lung cancer chemoprevention. SPIE,3260: 207–211, 1998.




2001;7:5-22. Clin Cancer Res Fred R. Hirsch, Wilbur A. Franklin, Adi F. Gazdar, et al. Recent Advances in Biology and RadiologyEarly Detection of Lung Cancer: Clinical Perspectives of

Updated version

http://clincancerres.aacrjournals.org/content/7/1/5

Access the most recent version of this article at:

Cited articles

http://clincancerres.aacrjournals.org/content/7/1/5.full#ref-list-1

This article cites 148 articles, 39 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/7/1/5.full#related-urls

This article has been cited by 35 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/7/1/5To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/content/7/1/5.full#ref-list-1

http://clincancerres.aacrjournals.org/content/7/1/5.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



early detection of lung cancer: clinical perspectives of ... · manifest lung cancer toward...

Documents